Full text of "Translated Texts for Historians"

See other formats

Second Revised Edition

Vegetius: Epitome
of Military Science

Translated with notes and introduction by
N. P. MILNER

LIVERPOOL

UNIVERSITY

PRESS

TRANSLATED TEXTS FOR HISTORIANS

This series is designed to meet the needs of students of ancient and
medieval history and others who wish to broaden their study by reading
source material, but whose knowledge of Latin or Greek is not sufficient
to allow them to do so in the original language. Many important Late
Imperial and Dark Age texts are currently unavailable in translation and it
is hoped that TTH will help to fill this gap and to complement the
secondary literature in English which already exists. The series relates
principally to the period 300-800 AD and includes Late Imperial, Greek,
Byzantine and Syriac texts as well as source books illustrating a particular
period or theme. Each volume is a self-contained scholarly edition
including an introductory essay on the text and its author. Notes on the
text indicate major problems of interpretation, including textual
difficulties. Translations are by individual specialists in the field.

General Editors:

Gillian Clark, University of Bristol
Mary Whitby, University of Liverpool

Editorial Committee:

Sebastian Brock, Oriental Institute, University of Oxford

Averil Cameron, Keble College, Oxford

Henry Chadwick, Oxford

John Davies, University of Liverpool

Carlotta Dionisotti, King’s College, London

Peter Heather, University College, London

Mark Humphries, National University of Ireland, Maynooth

William E. Klingshim, The Catholic University of America

Michael Lapidge, Clare College, Cambridge

Robert Markus, University of Nottingham

John Matthews, Yale University

Claudia Rapp, University of California, Los Angeles

Raymond Van Dam, University of Michigan

Michael Whitby, University of Warwick

Ian Wood, University of Leeds

Cover illustration: A sea fight (after a manuscript illumination, Vegetius
De Re Militari, fB6, Fitzwilliam Museum, Cambridge)

A full list of titles in the Translated Texts for Historians series is
available on request. Recent titles are shown below.

Bede: The Reckoning of Time

Translated with notes and introduction by FAITH WALLIS

Vol. 29, 352pp. including 7 line diagrams, 1999, ISBN 0-85323-693-3

Ruricius of Limoges and Friends: A Collection of Letters from
Visigothic Gaul

Translated with notes and introduction by RALPH W. MATHISEN
Vol. 30, 272pp., including 10 illustrations from the Codex Sangallensis
190, 1998, ISBN 0-85323-703-4

The Armenian History attributed to Sebeos

Translated with notes by ROBERT THOMSON

Vol. 31 (2-volume set), 240pp., 224pp., 1999, ISBN 0-85323-564-3

The Chronicle of Pseudo-Joshua the Stylite

Translated with notes and introduction by FRANK R. TROMBLEY and
JOHN W. WATT

Vol. 32, 240pp., including 3 maps and 2 plans, 2000, ISBN 0-85323-585-6

The Ecclesiastical History of Evagrius Scholasticus
Translated with an introduction and notes by MICHAEL WHITBY
Vol. 33, 454pp., 2000, ISBN 0-85323-605-4

Antioch as a Centre of Hellenic Culture, as Observed by Libanius

Translated with an introduction and notes by A. F. NORMAN
Vol. 34, 224pp., 2000, ISBN 0-85323-595-3

Neoplatonic Saints: The Lives of Plotinus and Proclus by their
Students

Translated with an introduction and notes by MARK EDWARDS
Vol. 35, 224pp., 2000, ISBN 0-85323-615-1

For full details, please write to the following:

All countries except the USA and Canada: Liverpool University Press,
4 Cambridge Street, Liverpool L69 7ZU, UK (Tel +44-[0] 151-794-
2233, Fax +44 [0] 151-794-2235, email J.M.Smith@llv.ac.uK

http://www.liverpool-unipress.co.uk) . USA and Canada: University of
Pennsylvania Press, 4200 Pine Street, Philadelphia, PA 19104-6097, USA
(Tel 215-898-6264, Fax 215-898-0404).

Translated Texts for Historians
Volume 16

Vegetius: Epitome
of Military Science

Translated with notes and introduction by
N. P. MILNER

Liverpool

University

Press

First published 1993

Revised second edition published 1996

Reprinted 2001

Liverpool University Press,

4 Cambridge Street
Liverpool, L69 7ZU

All rights reserved. No part of this
book may be reproduced in any form
without permission in writing from the
publishers, except by a reviewer in
connection with a review for inclusion
in a magazine or newspaper.

British Library Cataloguing-in-Publication Data
A British Library CIP Record is available

ISBN 0-85323-910-X

The publishers gratefully acknowledge the following
generous grants in aid of the publication of this book

Christ Church, Oxford

The Jowett Copyright Trustees

The Seven Pillars of Wisdom Trust

Printed in the European Community by
Bookcraft Limited, Midsomer Norton, Wilts

Contents

Preface.

Introduction.

§ 1. The work.

§2. The author.

§3. The date .

§4. The late-Roman army

. . xi
•««

. XIII

•««

. Xlll

. xxxi
xxxvii
. xlii

Ancient Synopsis.1

Book I.1

Preface.1

(Recruitment, ch. 1-7).2

1. That the Romans conquered all peoples solely because of their

military training.2

2. From what regions recruits should be levied.3

3. Whether recruits from the country or from the city are more useful.
.4

4. At what age recruits should be approved.5

5. At what height recruits should be approved.6

6. That the potentially better recruits are recognized at selection from

the face and physical posture.6

7. Of what trades recruits should be selected or rejected.7

(Training, ch. 8-28).9

8. When recruits should be marked.9

9. Recruits should be trained in the military step, in running and in

jumping. 10

10. Recruits should be trained in the art of swimming.11

11. How the ancients trained recruits with wicker shields and with

posts.12

12. Recruits should be taught to strike not with the edge, but with the

point. 13

13. Recruits should be taught armatura .13

14. Recruits should be trained at throwing javelins. 15

15. Recruits should be thoroughly trained in arrows.15

16. Recruits should be trained at firing stones from slings.16

17. On training with lead-weighted darts.17

18. How recruits should be trained in mounting horses.18

19. Recruits should be trained in carrying a burden.18

20. What kind of arms and armour the ancients used.19

21. On building camps.23

22. In what kind of places camps should be built.23

23. In what shape camps should be marked out.24

24. The sorts of fortification of a camp.24

VEGETIUS’ EPITOME OF MILITARY SCIENCE

25. How a camp should be built when under threat from the enemy.
.25

26. How recruits are trained to keep ranks and intervals in the line. 25

27. The distance that they should advance and retire, and how often a

month the exercise should take place, whereby soldiers are led out on
marching manoeuvres.26

28. On encouraging military science and Roman valour.27

Book II .29

Preface.29

(The armed services, ch. 1-3).29

1. Into how many categories military science is divided.29

2. The difference between legions and auxilia .31

3. What caused the legions to become depleted.32

(The formation and structure of the Ancient Legion, ch. 4-14) 34

4. How many legions the ancient generals led to war.34

5. How the legion is formed.34

6. How many cohorts there are in a legion, and how many soldiers in

a cohort.35

7. The titles and grades of officers of the legion.36

8. The titles of the officers who led the ancient centuries.40

9. The duties of the Prefect of the legion.41

10. The duties of the Prefect of the camp.42

11. The duties of the Prefect of engineers.43

12. The duties of the Tribune of soldiers.43

13. The centuries and ensigns of infantry.44

14. The troops of legionary cavalry.45

(The Ancient Legion in battle-order, ch. 15-18).47

15. How the lines of the legions are drawn up.47

16. How triarii and centurions were armed.49

17. When battle commences the heavy armament stands like a wall.
.49

18. The names and ranks of soldiers are to be written on the face of

their shields.50

(The administration of the Ancient Legion, ch. 19-25).51

19. As well as physical strength, skills in short-hand writing and

computation are required of recruits.51

20. Soldiers ought to deposit half their donative “with the standards”,

to be kept in savings for them.54

21. Promotions in the legion are made in such a way that those

promoted pass through all cohorts.55

CONTENTS

Vll

22. The difference between trumpeters, homblowers and the bugle

call.36

23. The training of soldiers.57

24. Examples to encourage military exercises, drawn from other arts.
.59

25. Catalogue of the legion’s tools and machines.59

Book III.62

Preface.62

(Logistics, commissariat, discipline, signals, castrametation, ch.

1-^).63

1. The proper size of an army.63

2. How the army’s health is controlled.65

3. How much attention should be devoted to the procurement and

storage of fodder and grain.67

4. Measures needed to ensure that soldiers do not mutiny.69

5. How many kinds of military signals there are.71

6. The degree of caution to be observed when an army moves in the

vicinity of the enemy.73

7. How to cross large rivers.78

8. How to lay out a camp.79

(Pre-battle strategy, ch. 9-13) .83

9. What and how many things are to be considered when judging

whether to engage the enemy in raids and ambushes or else in pitched
battle.83

10. What to do if one has an army unaccustomed to fighting or newly

recruited.86

11. Precautions to be taken on the day of engaging in a general

action.90

12. One should find out how soldiers are feeling before battle. . . 92

13. How a suitable place is chosen for battle.92

(Battle tactics and strategies, ch. 14-26).93

14. How the line should be drawn up to render it invincible in battle.
.93

15. Counting the square-footage, how much space in the line should be
left abreast between each man, and in depth between each rank. . 96

16. On deploying cavalry.98

17. On reserves, which are posted behind the line.99

18. In what position the commander-in-chief, the second- and third-in-

command should stand. 100

viii VEGETIUS’ EPITOME OF MILITARY SCIENCE

19. Remedies to counter the strength and stratagems of the enemy in

battle. 102

20. How many modes for engaging in a pitched battle there are, and
how the side that is inferior in numbers and strength may prevail. 104

21. An escape-route should be offered to the enemy so that they may be

more easily destroyed in full flight. . 107

22. How to retreat from the enemy if the plan to fight is rejected. 108

23. On the camels and armoured cavalry. 110

24. How scythed chariots and elephants may be resisted in battle. 112

25. What to do, if part or if all of the army is routed. 114

26. General rules of war. 116

Book IV. 120

Preface. 120

(Fortifications, ch. 1-6). 121

1. Cities should be fortified either by nature or by works. 121

2. Walls should be built not straight, but with angles. 121

3. How a terreplein is attached to the wall from ground-level. . . 122

4. On portcullises and gates, protecting them from fire. 122

5. On making fosses. 122

6. On preventing men on the walls from being injured by enemy

arrows. 123

(Preparations for a siege, ch. 7-11). 123

7. Measures to be taken so that the besieged do not suffer famine. 123

8. What supplies should be procured for the defence of walls. . . 124

9. What to do if the supply of sinews runs out. 125

10. What to do to ensure that the besieged suffer no shortage of water.

11. If salt runs out. 127

(Siege-strategies of attack and defence, ch. 12-30). 127

12. What to do when the first assault is pressed to the walls. . . 127

13. List of machines used in storming walls. 127

14. The ram, siege-hook and “tortoise”. 127

15. “Vines”, screens and the siege-mound. 128

16. Mantelets. 129

17. Mobile towers. 130

18. How a mobile tower may be fired. 131

19. How height is added to the walls. 131

20. How the ground may be undermined so that the machine can be

rendered harmless. 132

21. Ladders, drawbridge, "thrust-bridge” and swing-beam.132

CONTENTS ix

22. Catapults, mangonels, “scorpions”, crossbows, “sling-staves” and

slings, torsion-engines by which the wall is defended. 133

23. Mattresses, nooses, grapnels and heavy columns are useful against

rams.. 135

24. On saps, whereby the wall is undermined or the city penetrated.

25. What the besieged should do if the enemy break into the city. 136

26. What precautions should be taken to prevent the enemy occupying

the wall by stratagem. 136

27. When surprise attacks are made on the besieged. 137

28. What the besiegers do to avoid suffering surprise attacks from the

besieged.. 138

29. The type of torsion-engines with which a city is defended. . 138

30. How the measurement is obtained for making ladders and

machines. 139

(The Navy, ch. 31-33). 140

31. Precepts of naval warfare. 140

32. The titles of the officers commanding the fleet. 141

33. How warships got their name. 141

(Shipbuilding, ch. 34-37). 142

34. The care with which warships are built. 142

35. The astronomical observations according to which timber should be

cut. 142

36. The months in which timber should be cut. 143

37. The size of warships. 143

(The art of navigation, ch, 38-43). 144

38. The names and number of the winds. 144

39. In which months it is safe to sail. 146

40. How the signs of storms should be astronomically observed. 147

41. On signs of the weather. 147

42. On tidal waters, that is, ebb and flow. 148

43. On navigational knowledge and oarsmen. 149

(Naval warfare, ch. 44-^6). 149

44. On naval weapons and torsion-engines. 149

45. How ambushes are set in naval warfare. 150

46. What to do when naval battle is joined in an open engagement.
. 151

Abbreviations and Bibliography. 153

Select Bibliography. 154

Index of Gods, People and Places. 158

Preface

This work had its genesis, appropriately enough, at an excavation of the
Roman fort at Strageath, Perthshire, in September, 1983, where 1 was
shown a copy of Vegetius’ Epitome by Mr. N. Puentes. It was due to
his encouragement in the early stages that I embarked on the project of
translation. Difficulties of interpretation along the way led to my
returning to Oxford to research the topic for a D.Phil. Some of the main
results of this research are incorporated here.

The technical nature of the original implies that many terms ought not
to be interchangeable, and had a precise meaning for Romans which is
not always easily grasped in English today. However, the translator’s
task is not helped by the fact that Vegetius himself was using them as
a layman and did frequently interchange them or use them imprecisely.
Furthermore, his interests were markedly philological; etymologies play
a large part in the author’s interpretation of terms. That being so, I have
thought it best to adopt a fairly literal style of translation, sticking close
to Latin terminology and word order, and retaining a number of Latin
technical terms. Quotations by Vegetius of other authors are given in
my own translations.

The rationale behind the footnotes has been first to explain my
decisions at textual cruces, to incorporate the most relevant philological
comments of previous scholars, to explain references to historical
persons and events, to provide cross-references to repeated ideas or
passages, to quote or cite relevant examples of language or ideas from
other ancient authors, and to highlight the nature of Vegetius’ method
of writing as an epitomator intent on shaping and imposing a decided
slant on his material. References to scholarly articles and works are
provided with the aim of directing interested readers to further research
if they so wish. When an ancient text has been commented upon so
rarely as this has, much of what is said here must be taken as only
provisional. It is intended here merely to provide an instrument for
further study by others.

Most of the first edition was read in draft by Dr. R.S.O. Tomlin and
Dr. Michael Whitby, to whom I remain greatly indebted for numerous
improvements. For this second edition I have refined and revised the
translation, augmented the notes, and expanded the introduction and
bibliography. I am grateful to Dr. Gillian Clark for much help with the
conversion of the disk and other editorial matters.

Introduction

§1. The work

The Epitoma Rei Militaris or Epitome of Military Science by Publius
Flavius Vegetius Renatus was in the Middle Ages one of the most
popular Latin technical works from Antiquity, rivalling the elder Pliny’s
Natural History in the number of surviving copies dating from before
AD 1300. A number of early translations into vernacular languages were
made, and frequently additions and adjustments were introduced to adapt
the work to the age of Chivalry. Sections of it were often reproduced
to augment more contemporary material on warfare, and in particular
the “General Rules of Warfare” (III.26) were found suitable for
repetition as inculcating the basic principles in an unspecific form which
could be adapted to serve a great variety of military situations.'

In the Renaissance high esteem continued to be shown, so that it
appeared along with Frontinus’ Stratagems^ Aelian’s Tactics and
“Modestus’ On terms of military science"' in several printed collections
of Roman militaria already in the century from 1487. “Modestus” is
itself a testament to the hunger for such material, being a
pseudepigraphical abridgement of Vegetius’ Epitome which was
published in Venice in 1471 and was long regarded as one of Vegetius’
sources.^ Even now it is sometimes taken as genuine,^

However, it seems that one can also have too much of a good thing.
Learned commentators such as G. Stewechius (Leiden 1585) expressed
disappointment that it was due to the preservation of Vegetius that
works by his named sources, Cato, Celsus, Frontinus and Patemus, had
not survived"^ whereas Vegetius’ Epitome, a late Christian source not
from the best period of Roman culture, gave no indication which parts
were owed to which classical author. This complaint is almost certainly
misguided; there are a number of indications in the Epitome that even
Vegetius worked from late epitomes of the named sources, so that it

'Cf. M. Springer, “Vegetius im Mittelalter”, Phiiologus cxxiii (1979) 85-90; F.H.
Sherwood, “Studies in medieval uses of Vegetius’ Epitoma Rei Militaris", Diss. Univ. of
California (Lx)s Angeles 1980), resume in DA xli (1980) 1712A; C.R. Schrader, “A
handlist of extant mss. containing the De re militari of FI. Renatus Vegetius”, Scriptorium
xxxiii (1979) 280-305; id., “The influence of Vegetius’ De re militari". Military Affairs
xlv (1981) 167-172.

^Cf. M. Jahns, Geschichte der Kriegswissenschaften (Munich 1889-91) 1.122, A.R.
Neumann, RE Suppl. X (1965), s.v. “Vegetius”, col. 992-1020 at 997.

^e.g. by B. R6my, “Notes de Lecture: Soldats de Cilicie sous !’Empire romain”, EA x
(1987) 107-109, cf. SEG xxxvii (1987) 1760.

^Reprinted in N. Schwebel, Flavii Vegetii Renati comitis de Re Militari libri quinque
(Nuremberg 1767), augmented edn. Strasbourg, 1806, p. vi (Stewechius’s dedicatio,
Leiden 1584).

XIV

VEGETIUS’ EPITOME OF MILITARY SCIENCE

may be doubted that the latter would have had an independent chance
of survival.‘ Secondly, in spite of the demand for Roman military
treatises from the Carolingian period onwards, no manuscript of the lost
authors can be demonstrated to have been discovered.

Vegetius’ reputation further suffered by comparison with ancient
historians such as Polybius in the attempt to piece together Roman
military history. N. Machiavelli’s L ar/e della Guerra (Florence 1521),
a thoroughgoing attempt to augment, modernize, illustrate and
supplement Vegetius in the light of all the evidence of classical warfare
available to him, made notable use of Polybius, Frontinus and Livy. P.
Scriverius certainly regarded Vegetius as a poor substitute for the
Polybius,^ after J. Lipsius’ De militia Romana: Commentarius ad
Polybium (Antwerp 1596) had accused Vegetius of confusing the
institutions of diverse periods of the Roman Empire,^ an accusation
which stuck. A long period of deepening neglect followed. Fewer
printed editions of Vegetius appeared in the seventeenth century, and no
new English translation was published after Lt. John Clarke’s Military
Institutions of Vegetius (London 1767), although T.R. Phillips reprinted
parts of Clarke’s in 1940 and 1944,until L.F. Stelten’s edition of
1990.^

Few commentaries have appeared. N. Schwebel’s (Nuremberg 1767,
Strasbourg 1806) was the last complete one of importance to ancient
historians; D. Nisard’s largely transmits the work of his predecessors.^
Count Turpin de Criss^’s Commentaires (Montargis 1775) are more
about eighteenth-century France and its relations to antique warfare in
general than an attempt to analyse the Epitome on its own terms. In the

'N,P. Milner, Vegetius and the Anonymus De Rebus Bellicis, D.Phil. thesis (Oxford 1991)
ch. 7. Most of the arguments in this introduction are set forth more fully in the thesis.

^Sic prod Penelopes, cum ad ipsam dominam accessus non pateret, cum ancillis illius
miscebantur, reprinted in Schwebel (1806) p. xii (from Scriverius’ praefatio, Leiden
1632).

^R. Sablayrolles, “Bibliographie snr Vepitoma rei militaris de V6g6ce”, CGRAR iii (1984)
142.

^T.R. Phillips (ed.), Flavius Vegetius Renatus, the Military Institutions of the Romans
(Harrisburg, Pa. 1944), reprinted Westport 1985; T.R. Phillips (ed.), The Roots of
Strategy: a collection of military classics (Harrisburg, Pa. 1940).

^L.F. Stelten, Flavius Vegetius Renatus: Epitoma Rei Militaris, edited with an English
translation (New York, Bern, Frarikfurt, Paris 1990).

‘’D. Nisard (ed.), Ammien Marcellin, Jornandes, Frontin, Vegece, Modestus, avec la
traduction en frangais, Coll, des Auteurs Latins 11.15 (Paris 1878). The translation was
reprinted by F. Reyniers, Vegece (Paris 1948).

INTRODUCTION

twentieth century, however, D.K. Silhanek produced a small unpublished
commentary on books I and II, ‘ and D. Schenk published a monograph
of Quellenforschung.^

Work on the critical edition of the text of the Epitome has been less
neglected. Since Schwebel, C. Lang edited the Teubner editions of 1869
and 1885, A. Andersson published a philological thesis on the text,^ and
L.F. Stelten produced his edition mentioned earlier, after unpublished
work on books I and 11."* New discoveries of manuscripts have for some
time made new critical editions desirable, even if initial indications are
that the new MSS belong to the family 7i whose authority Lang
normally preferred to follow,^ and another new edition was
reported in progress for the Bud^ series in 1977.^ There is also a new
Teubner edition by A. Onnerfors which appeared in 1995.^ The present
translation is from the text of Lang’s second edition, except a few cases
where I have adopted a variant reading from Lang’s apparatus, as
indicated in the notes. I have normally not transmitted Lang’s square
brackets, indicating text which he considered to be interpolated by
scribes, because the work of classical philologists such as M. Schanz,®
A. Andersson and P. De Jonge^ has all tended in the direction of
vindicating the suspected text, except in a very few cases indicated in
the notes.

Scholarly interest in Vegetius has slowly been rising since the
nineteenth century when a number of German dissertations were devoted
to the Epitome. If interwar activity in this area by E. Sander, F.

'D.K. Silhanek, A translation and commentary on Vegetius ’ Epitoma Books land 11, Diss.
(New York Univ. 1972).

^D. Schenk, Flavius Vegetius Renatus: Die Quellen der Epitoma rei militaris, Klio,
Beitrftge zur alten Geschichte xxii (N.F, ix) (Leipzig 1930), repr. Nuremberg 1963.

^A. Andersson, Studia Vegetiana, Diss. (Uppsala 1938).

^L.F. Stelten, Epitoma Rei Militaris of Flavius Vegetius Renatus. A critical edition of
books I and II, Diss. (St. Louis Univ. 1970), r6sum6 in DA xxxi (1970) 2363A.

^Cf. C.E, Finch, ‘‘Codices Pal. Lat. 1571-1573 as sources for Vegetius”, TAPhA xciii
(1962) 22-29, id., “Source of codex E of Vegetius”, Classical Bulletin xli (1965) 45—46,
L. Rubio, “El ms. Scorialensis L.II1.33: Nuevos dates para una futura edicion del Epitoma
Rei Militaris de Vegetius”, Emerita xli (1973) 209-223.

'^W. Goffart, “The Date and Purpose of Vegetius’ De Re MilitarF, Traditio xxxiii (1977)
65 = id., Rome’s Fall and After (London 1989) 45 n.l, cf. addendum p. 355.

’But see the review by M.D. Reeve, RFIC (forthcoming); and R.S.O, Tomlin, Gnomon
(forthcoming).

®M. Schanz, “Zu den Quellen des Vegetius”, Hermes xvi (1881) 137-146.

’P. de Jonge, “Ammianus and Vegetius”, in Vt pictura poesis: studia Latina Petro
Johanni Enk septuagenario oblata (Leiden 1955) 99-106.

XVI

VEGETIUS’ EPITOME OF MILITARY SCIENCE

Lammert and A.R. Neumann was not sustained post-1945, at least the
results of a century of German Vegetian scholarship could be collected
in Neumann’s RE article of 1965,’ Since then seminal articles by F.
Paschoud,^ W. Goffart^ and G. Sabbah'* have made considerable
advances by considering Vegetius in the context of his own times,
themselves greatly illuminated by the rapid growth of the study of late
Antiquity.

This translation is made on the principle that since Vegetius addressed
his work to a late-Roman Emperor, it is only by trying to understand
what it could have meant to him that we can begin to evaluate it. To
feel disappointment that Vegetius was not writing history is a symptom
of having set out with the wrong preconceptions as to the nature of the
work. Close examination reveals that it is highly topical, by way of
offering a systematized remedy for alleged military failures in
recruitment and training, army organization and strategy, and arms and
equipment. It is also very selective; cavalry warfare and river patrol-
boats are expressly omitted on the grounds that late-Roman progress in
these areas rendered them above criticism; attacking hostile cities is
largely left out presumably because the main contemporary barbarian
enemies, the Goths, Huns and Alans, did not live in cities. The use of
barbarian auxiliaries of various kinds is generally ignored for reasons
probably determined by Vegetius’ programme to apply the legionary
model in the main. The late-Roman two-tier military organization of
comitatenses and palatini in the field armies, and limitanei and ripenses,
cohortes and alae, in the frontier armies is hardly visible, because
Vegetius was mainly interested in the field armies.

The work also falls self-evidently into the genre of epitomes of
technical treatises. As there were different types of epitome, it is
important to establish which type our Epitome is, namely whether it is
a faithful summary of other works, preserving their general order and
arrangement, or whether it is a “scissors-and-paste” mosaic of other
works excerpted and rearranged according to the epitomator’s own
system, and if the latter, whether it incorporates much additional

’A.R. Neumann, RE Suppl. X (1965), s.v. “Vegetius”, col. 992-1020.

^F. Paschoud, Roma Aeterna, Bibliotheca Helvetica Romana VII (Neuchatel 1967)
110-117.

^See above.

^G. Sabbah, “Pour la datation th^odosienne du De Re Militari de Veg^ce”, Centre Jean
Pateme, Memoires II (Univ. de Saint-Etienne 1980) 131-155.

INTRODUCTION

xvii

material of the epitomator’s own composition. Close inspection has
produced opposite conclusions. Whereas Schenk thought that each of the
four books was largely based on a single source—book 1 on Celsus, II
on Patemus, and III-IV on Frontinus—it may be argued that although
some signs of Frontinus’ arrangement may be discerned in III and IV
the work is on the whole drafted after the “scissors-and-paste” model
and is extensively augmented and interpreted by the epitomator.

So much so, indeed, that in book I and 11 there seem grounds for
thinking that Vegetius greatly expanded jejune source-material on the
training and organization of the “ancient legion”, using methods of
subdivision of subjects, repetition of material, antiquarian etymologizing,
modem glossing, embellishment with literary allusions and historical
exempla, provision of medical explanations of phenomena and Christian
pieties, reconstruction of ancient institutions from intelligent guesswork
sometimes based apparently on contemporary civil service practice, and
choice of material and emphasis determined by the wish to further a
specific programme. The same methods may also be demonstrated for
parts of book III designed to promote strategies of seizure and
preservation of supplies and foodstuffs and the harrying of the enemy
invader by means of guerilla actions rather than pitched battles, and in
book IV in the accent placed on the defence of cities and supplies, and
on themes of heavy armour and ambush pursued perhaps rather
incongrously in the account of naval warfare.

In spite of all this, there are some convincing parallels between parts
of book III and fragments of Cato de Re Mi I i tar i preserved in other
works, and the battle-order of the “ancient legion” retains the obsolete
technical terms for the categories of combatants— principes, has tat i,
triarii, velites, ferentarii, accensi —normal in the third and second
centuries BC. The background to the accounts of recruitment and
training in books I, II and III also seems to link up with Cato, both in
literary parallel passages and in the historical situation after the Second
Punic War mentioned in significant chapters of books I and III.
Furthermore the strikingly anomalous use of the verbal 2nd. pers.
singular and tuus-a-um in book III is convincingly ascribed to the

xviii VEGETIUS’ EPITOME OF MILITARY SCIENCE

influence of Cato’s original words/ and some of the tactics there
described can be identified with battles of the Second Punic War/

If the basic substrate of the “ancient legion” is Cato, many of the
details of organization and administration in book II are hardly so
ancient. Some seem recognizably parallel to late-Roman military and
civil service grades, such as supernumerarii, decani, administrative
centuriones - centenarii, and a princeps in overall charge of
administration, and there is an unparalleled concatenation of imperial
praefectus legionis, praefectus castrorum and tribuni vel praepositi all
in command of the same unit with duties ambiguously differentiated.
Some of the other details may be explained as reconstructions from
obsolete terminology, such as the century of 100 men and the praefectus
fabrorum/fabrum arbitrarily put in charge of legionary technicians and
engineers.^ The Vegetian cohort of five centuries may have been
influenced by decimal subdivisions of late-Roman units.

Vegetius provides two source-notices, a fuller one in 1.8 citing Cato
the Censor, Cornelius Celsus, Frontinus, Patemus and the “constitutions
of Augustus, Trajan and Hadrian”, and a summary one in 11.3 which
emphasizes “Cato the Elder” and Frontinus among alii
conplures —“many others”. There is no reason to think that a different
set of sources is meant for books II-IV, which came out after book I
and so have a separate source-notice.

Book I, about the recruitment and training of legionaries, promises
because of the subject and the frequent mention of his name to rely
largely on Cato’s de Re Militari. But in the laborious circumstances of
ancient book production, it was normal for writers to cite authorities at
second- or third-hand or even further removes. Caesar, for instance,
named the antique Eratosthenes rather than Poseidonios whom he
actually followed.^ As there was a common ancient tendency to suppress

'D. Schenk, 60-61. Cf. V. Epit. 111.11 note.

^The battle of llipa in 206 BC: V. Epit. III.20 Lang pp.l08.20-I09.17, cf. Livy 28.14,
Polyb. 11.20-24, Sen. de Vit. Beat. 4.1; see M. J^ns, 54-55. The counter-tactic to
elephants at the battle of Zama, 202 BC: V. Epit. III.24 Lang p. 118.4-8, cf. Livy
30.33.1-3. The battle of Cannae, 216 BC: V. Epit. 111.14, Floras 1.22.16, cf 1.38.15, C.
Marius later used Hannibal’s “art of Cannae” Xo defeat the Cimbri.

^Cf. D. Baatz, Bauten und Katapulte des romischen Heeres, Mavors XI (Stuttgart 1994)
78, 133.

'*A. Klotz, Casarstudien (1910) 27, cited by F. Lammert, Gnomon x (1934) 271-274.

INTRODUCTION

XIX

the names of the more immediate sources,' or to copy authors
unnamed,^ we have to reckon with the possibility that Vegetius may
have actually used epitomes of even Frontinus and Patemus. This would
seem to be indicated for instance if there were signs that he had a
persistent problem of too little information to go on.

For most of those who cited Cato’s de Re Militari, Verrius Flaccus
is the likely source.^ Festus’, Nonius Marcellus’, and Priscian’s citations
of linguistic peculiarities certainly or probably derive from the Augustan
scholar’s grammatical compilations de Obscuris Catonis or de
Significatu Verborum^ Because of the archaic word vitiligunt, the elder
Pliny’s quotation in the preface to his Natural History looks suspiciously
like a fragment preserved in such a source too.^ Apart from Gellius,
Vegetius is the only other citator, but if he had really had access to the
original work he would surely have made more of the fact. He never
explicitly quotes Cato verbatim, and in the one case where we can
compare the original words, it is not clear that he knew he was quoting
Cato, and his version shows linguistic modernization of the original.^ It
is obvious that if he had had the original text, he would have had to
interpret dozens of archaisms, but of this, not a word is said. A work
which was probably a rarity even four hundred years earlier, and which
no one can be demonstrated to have seen after Gel Hus in the second
century, was surely not in Vegetius’ hands.

Moreover, Schenk’s analysis of Epitome Ill suggests that “Frontinus”
subsumed “Cato”, at least (see below), and the fact that Celsus and
Frontinus are presented in a zeugma in the source-notice at 1.8, coupled
with the fact that Celsus was writing under Tiberius between Cato and
Frontinus, probably means that we should assume that “Frontinus”
subsumed “Celsus” too. Whether or not the plan of A. Cornelius Celsus’

'Cf. Jerome’s habitual citation and even quotation of Greek authors ostensibly from the
originals but actually from Latin secondary literature, translations, handbooks,
commentaries, etc., P. Courcelle, Late Latin Writers and their Greek Sources (Cambridge,
Mass. 1969) 78 ff.

^Pliny Hist. Nat. praef. 21 ff.

'AuIus Gellius 6.4.5 is the most probable exception; this second century scholar had a
strong liking for Cato’s works, and took the trouble to hunt them out. Even so, he cites
de Re Militari only once.

'‘Cf. Festus 306 M.; D. Schenk, 36-37.

^Pliny Hist. Nat. praef 30, cf 32, remarking Cato’s compound vitilitigatores. For all the
citations, cf Jordan pp.80-82.

^'Epit. III.20 note.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

largely lost Encyclopaedia containing four artes —agriculture, medicine,
rhetoric and military science—was inspired by Cato’s works, which
included his de Agricultura and de Re Militari, it is likely that Celsus
transmitted the Catonian material on the army to Frontinus.' Schenk
argued that Celsus was the main source for book I on stylistic grounds
that have fallen victim to Andersson’s demonstration that the style of
the whole Epitome is in general (apart from the use of the 2nd. pers.
sing, in book III) homogeneous.^

If Frontinus borrowed from Celsus, we have to ask also from what
source Josephus, writing his Jewish War in Rome in the AD 70s^
between Celsus and Frontinus, took his famous excursus on the Roman
army and its discipline, training, philosophy, equipment and camp-
construction, which has much in common with Epitome 1-II.'‘ Josephus’
information on castrametation and camp discipline also shows some
parallels to Vegetius’ material on castrametation.^ In two respects the
latter is less detailed than Josephus; he does not discuss the system of
the tessera that controlled the watches and sentries, and omits the
traditional method of striking camp.^ There is no good reason for these
omissions. Obsoleteness does not in general deter him, and the evidence
suggests anyway that the tessera still existed in the late fourth century
AD.^ Therefore Vegetius’ source was even more concise. If Josephus’
source was Celsus, then Vegetius did not make direct use of Celsus; still
less did he consult Cato. But the material may well have derived from

'F. Marx (ed.), A. Cornelii Celsi quae supersunt (Leipzig-Berlin 1915) vii-viii.

^D. Schenk, 28 ff,. A, Andersson, 21 ff.

Rajak, Josephus, the Historian and his Society (London 1983) 195. N.b. Josephus
could and did translate from Latin writers, cf. ibid. 235.

‘‘Jos. Bell. lud. 3.71, V. Epit. I.l, cf. Jos. Bell. lud. 2.577 (Roman Empire won by valour
not by fortune). Jos. Bell. lud. 3.72-75, V. Epit. LI, 1.4, 1.27-28, 11.23-24, III praef,
III.4, III.9 p.88.15 ff. (Lang), III. 10, cf IV.31-32 (Romans always training). Jos. Bell,
lud. 3.102, V. Epit. I.l, 1.2, 1.7,1.8,11.19 (training of minds as well as bodies). Jos. Bell,
lud. 3.88, 104-105, V. Epit. 11.2,11.21 (Roman army behaves as one body). Jos. Bell. lud.
3.106, V. Epit. 1.8, II.2 fin. (Roman invincibility in the stationary battle primarily
responsible for the size of their Empire).

^Jos. Bell. lud. 3.76, V. Epit. 1.21, III.8, III.IO p.92.1 Off (Romans always build a camp,
so are never vulnerable to surprise attack). Jos. Bell. lud. 3.78, V. Epit. II. 11 (Roman
army takes with it a multitude of artificers and builders for this purpose). Jos. Bell. lud.
3.79-84, V. Epit. 11.18, cf 11.25 fin. (camp “resembles a city”)- Jos. Bell. lud. 3.85, V.
Epit. 11.19, 111.8 {lignatio, pabulatio, aquatio).

'Jos. Bell. lud. 3.87-92, Polyb. 6.34.7-12, 6.40.1-3.

’Amm. 21.5.13; 23.2.2; O. Douch, an archive from a limitanean unit in Egypt c. AD 375:
tesserarius, nos. 12, 15, 30, 33, 41, 53; loh. Lyd. de Mag. 1.46.

INTRODUCTION

XXI

Celsus summarized in Frontinus or Patemus, or rather we should say
from an epitome of either or both.

Despite suggestive parallels, Josephus’ excursus probably did not
derive from Polybius’ excursus on the Roman military system preserved
in his Histories. ' For although, whilst omitting much archaic matter,^ he
also transmitted common details,^ where Polybius was brief, such as for
the method of striking camp, Josephus is much fuller, and where
Polybius dwelt at length, for instance on the savage punishments of
decimation and fustuarium, and on rewards for bravery, Josephus is
brief. Given the Roman subject, and the prescriptive nature of Polybius’
text, it seems likely that his account went back to a Latin model which
was more detailed still, and that Josephus’ account ultimately derived
from the same more detailed Latin model. We may infer from the
evidence of Vegetius-Josephus parallels that Josephus’ immediate source
was a presumably first-century AD military handbook such as that by
Celsus that derived ultimately from Cato’s de Re Militari and also that
the latter was Polybius’ main source, written by his elder contemporary

As for Sex. Julius Frontinus, who was consul ter or dinar ius with the
Emperor Trajan in AD 100, and is named both in 1.8, and more
particularly in 11.3 as Vegetius’ main source after Cato, we are
unfortunate in the loss of his de Re Militari. However his surviving
Strategemata, which is generally agreed to have furnished exempla to
the lost theoretical work,^ was analysed for parallels to the structure of

'Polyb. 6.19-6.42.

^e.g., Polyb. 6.19-26 (raising of consular legionary armies).

^Jos. Bell. lud. 3.77, cf. Polyb. 6.31.10 (square-shaped camp). Jos. Bell. fud. 3.83, Polyb.
6.31.10 (simile of a city). Jos. Bell. fud. 3.89-92, Polyb. 6.40.1-3 (three-signal method
of striking camp). Jos. Bell. fud. 3.102-104, Polyb. 6.37-38 (savage discipline). Jos. Bell,
fud. 3.87-88, Polyb. 6.34.7-12 (tessera). Jos. Bell. fud. 3.93-97, Polyb. 6.22-23 (infantry
arms), 25 (cavalry arms).

‘‘E. Rawson, PBSR xxxix (1971) 13 ff., argued that Polybius’ excursus derived from
commentarii of military tribunes, since it lays down the basics from their point of view.
But the same would not be surprising of Cato as they were the most senior officers of the
legion after the consul, who would naturally not be concerned with such details. The fact
that Cato-Vegetius diverges on the order principes-hastati-triarii from Polybius’
hastati-principes-triarii is without weight, given V.’s editorial method and remoteness
from the original. Since Polybius’ source dated from before his own day, but as Rawson
observes must postdate the introduction of the military oath in 216 BC (Livy 22.38, cf.
Front. Strat. 4.1.4), Cato as a generation older than Polybius would fit the profile. Finally,
Cato de Re Mil. fr. 8 (Festus 298 L., 253 M. = Jordan p.81) on procubitores probably lies
behind Polybius 6,35,5 on TtpoKoiTi'a,

^C. Wachsmuth, RhM xv (1860) 575, endorsed by R. Grosse, DLZ Iv (1934) 61-65.

xxn

VEGETIUS’ EPITOME OF MILITARY SCIENCE

Epitome III and IV by Schenk.' Although the structure of Frontinus’ lost
work remains uncertain, the Strategemata allows some deductions to be
made about it from the order of subjects and rubrics. They imply that
the Catonian tactical chapters of Epitome III.9-22 are embedded in the
part most likely to derive from Frontinus. But there are possible
exceptions to the Frontinian scheme in III. 1-3, 5, 8, 26 at least, not to
mention the naval chapters in book IV.

Because Frontinus may have been influenced by the Strategikos of
Onasander, a Greek philosopher writing in Rome during the mid-first
century AD, parallels between Onasander and Vegetius also support the
partial reconstruction of the structure and content of Frontinus’ work.^
Not only is the general pattern of pre-battle, battle and post-battle,
followed by the treatment of siege warfare, observable in Onasander as
in Epitome III-IV and Frontinus Strat, I-III, but there are also
numerous resemblances of detail between Onasander and Vegetius.^ The
order of subjects and the sheer number of connexions, many in addition
to those collected by Schenk, make it likely that Frontinus’ work
included from Onasander a number of the same points which have

‘D. Schenk, 44-81,

^E. Sander, Philologische Wochenschrift xWx (1929) 1230-1231, D. Schenk, 81-83, F.
Lammert, Philologische Wochenschrift lix (1939) 236-237.

^Onas. 6.8, V. Epit. 111.5 (dust-cloud and fire-signals). Onas. 6.9, V. Epit. III.11 (short
march to battle). Onas. 7.1-2, V. Epit. III.6 (occupying mountain passes). Onas. 7.2, V.
Epit. 111.2 (avoiding marshes). Onas. 9.1, V. Epit. III.2 (continual changes of camp). Onas.
9.2, V. Epit. 11.23 cf 1.18, 111.2 (continual drill in winter). Onas. 10.1-3, V. Epit. 1.26,
cf. 11.2, 111.9 (infantry manoeuvres). Onas. 10.4 (armed with staves), V. Epit. 1.11-12
(training with staves), 1.13, 11.23 {armatura), 111.9, III.4 (mock-battles). Onzis. 10.6, V.
Epit. 1,26, 111,2 (cavalry manoeuvres). Onas. 10.10-12, V. Epit. III.8 (guard by night in
relays). Onas. 10.13, V. Epit. 111.22 (night retreats). Onas. 10.15, V. Epit. 111.6 (guides).
Onas. 10.20, V. Epit. 111.11 (remaining in fortification and launching surprise attack).
Onas. 10.22-24, V. Epit. 111.6, cf. 111.26 (secrecy). Onas. 11.1-5, V. Epit. 111.6, 111,22,

III. 25 (pursuit in broken country). Onas. 12.1-2, V. Epit. III.11, cf Front. Strat. 2.1.1
(meal before battle). Onas. 13.1-3, V. Epit. 111.9, 111.12 (exhortation in adversity), Onas.
19.1, V. Epit. 1.20, 11.17,111.14 (intervals in the line for the light-armed). Onas. 21.1, V.
Epit. 111.15 (depth better than breadth). Onas. 21.3, V. Epit. 111.20 (protecting the flank
using natural features: 7th. depugnatio). Onas. 21.5-7, V. Epit. 111.20 (crescent formation:
4th. and 5th. depugnationes). Onas. 21.8, V. Epit. 111.20 (oblique formation: 2nd., 3rd.
and 6th. depugnationes). Onas. 22.1-4, V. Epit. III. 17-20 (tactical reserve). Onas. 26, V.
Epit. III.5 (gestures), Onas. 28-29.2, V. Epit. 1.20, II.12, 11.14 (shining armour). Onas.
31, V. Epit. 111.13 (choosing broken ground if enemy stronger in cavalry). Onas. 36.3-6,
V. Epit. 111.25 (exhortation in defeat). Onas. 38.7-8, V. Epit. III.6, 111.26 (reception of
defectors). Onas. 39.4-7, V. Epit. IV.12 (terrorizing cities). Onas. 40-41, V, Epit. IV.28
(night-sallies). Onas. 42.17, V. Epit. IV. 12 (trumpets in sieges). Onas. 42.18-21, V. Epit.

IV. 25 (advice to citizens to surrender arms). Onas. 42.23, cf V. Epit. IV.7 (sending non-
combatants into besieged city).

INTRODUCTION

XXlll

finally got into Epitome III-IV. But at no point is there a really close
verbal correspondence between Onasander, what survives of Frontinus,
and Vegetius. Even where the latter has exempla in common with
Frontinus the differences are problematic for any theory of direct use of
the Strategemata} It should be noted too that Vegetius, being on the
whole ignorant of Greek (see below), would not have used Onasander
directly.

The Greek origins of much of the form and substance of Epitome
III-IV ought not to be doubted, including matter on land tactics
mediated through Cato and Frontinus and the inclusion of a section on
naval warfare.^ What is far from clear is whether Frontinus included
sections on shipbuilding and navigation, for Epitome IV.34-42 seems to
carry material derived from the libri navales of the first century BC
polymath M. Terentius Varro, cited at IV.41. A change of source is
indicated both by an observable difference in treatment of the material,
and the fact that what can be deduced from surviving fragments of
Varro’s “naval books” in Pliny, Isidorus, and Servius “Fuldensis” (a
Vergilian commentator of the late fourth century ad) seems to lie
directly behind the chapters on navigation IV.38^2.^ That Varro was

'On V. Epit III.10 p.93.2-11 (Lang), Val. Max. 2.7.1-2, Front. Strat. 4.1-2, see Schenk,
48, 59, citing A. Klotz, “Exempla und Epitoma Livii”, Hermes xliv (1909) 211 ff. The
most parallel exemplum is Front. Strat. 4.7.16: Scipio AJricanus dicere solitus est hosti
non solum dandam esse viam adJugiendam, sedetiam muniendam, “Scipio Africanus used
to say that a way for the enemy to flee by should not only be given to them, but even
built for them,” and V. Epit. III.21: Ideoque Scipionis laudata sententia est, qui dixit viam
hostibus, quaJugerent, muniendam, “For this reason Scipio’s axiom has won praise, when
he said that a way should be built for the enemy to flee by.” The differences are as
striking as the similarities between Front. Strat. 4.7.27: Scipio Aemilianus ad Numantiam
omnibus non cohortibus tantum, sed centuriis sagittarios etfunditores interposuit, “Scipio
Aemilianus atNumantia interspersed archers and slingers not just in all cohorts but in (all)
centuries”, and V, Epit. 1.15: Africanus quidem Scipio, cum adversum Numantinos, qui
exercitus populi Romani sub iugum miserant, esset acie certaturus, aider se superiorem
futurum esse non credidit, nisi in omnibus centuriis lectos sagittarios miscuisset, “Scipio
Africanus, about to do battle with the Numantines who had sent armies of the Roman
People under the yoke, supposed that he would only get the upper hand if he incorporated
picked archers in all centuries.”

W. Epit. Ill praef., cf. IV.31 note; Aelian Tact praef 3 on Frontinus; Frontinus’
surviving Strategemata covers Greek military history as well as Roman, and naval in
addition to land warfare, but not separately.

^See IV,38, note.

xxiv VEGETIUS’ EPITOME OF MILITARY SCIENCE

also behind those on shipbuilding IV.34-37 is no more than a
possibility.’

In his fuller source-notice at 1.8 Vegetius also cites “Paternus, a most
zealous champion of military law”, and the “constitutions of Augustus,
Trajan and Hadrian”. Contrary to the view of Alfred Neumann and
others that he used directly or indirectly the rule-book of Hadrian, which
subsumed that of Augustus, itself a compilation based on Republican
rule-books conjecturally going back to P. Rutilius Rufus, Scipio
Africanus Minor and Cato de Re Militant Vegetius seems aware only
of ad hoc military regulations.^ Like Schanz and Grosse, Schenk felt
that the constitutiones were available to him only in so far as they were
cited by the Antonine writer and praetorian prefect P. Taruttienus
Patemus, and this seems preferable.'’

Surviving fragments of Patemus’ probably Juristic work in Justinian’s
Digest and John the Lydian’s de Magistratibus show that he covered in
some detail in four books^ both antiquities of the Roman army in the
age of, for example, Romulus and current regulations of the second
century AD, supported by quotations from the disciplina Augusti.^
Schenk, drawing attention to an alleged similarity between the Digest
and Epitome 11.19, and to Vegetius’ announcement that he would
proceed to explain the organization of the ancient legion “following the
guidance of military law” (11.4), thought that book 11 was based mainly

'The astronomical observations for felling and hewing timber occur also in Pliny Hist.
Nat. 16.74.190-191 and Servius ad Verg. Georg. 1.256, cf V. Epit. IV.35 note; Pliny
16.18.42 says that the fir —abies —was in great demand for building ships. The further use
of material deriving from Varro’s works by V. may also be suggested for 1.2 {de
Architectural) and 1.6 {Rer. Rust.).

^A. Neumann, CPh xxxi (1936) 6-7, 9; id. CPh xli (1946) 223. A.A. Schiller, “Sententiae
Hadrianae de re militari”, in Sein und Werden im Recht: Festgabe fur Ulrich von Liibtow
zum 70. Geburtstag am 21. August 1970 (Berlin 1970) 295-306, argues that there may
well have been no rule-book as such.

^V. Epit. 1.27 on marching drill, where he omits the name of Trajan. There is further
scope in book I for use of such regulations in 1.9 on marching drill and in 1.5 on tlie
height of recruits, and perhaps in 11.22 on military executions; elsewhere their use seems
improbable.

'‘D. Schenk, 12 ff. See V. Epit. 1.8 note.

^Dig. index Florentinus.

*Cf. loh. Lydus de Mag. 1.9 (Romulus), Dig. 50.6.7 (long list of second century AD
immunes\ Dig. 49.16.7 (precept on capital punishment of proditores and transfugae). Dig.
49.16.12 §1 (precept forbidding use of soldiers for private services, quoting from the
disciplina Augusti, cited at V. Epit. 11.19 note).

INTRODUCTION

XXV

on PatemusJ But it was in 1.8, the source-notice for book I which was
originally published separately, that Vegetius named Patemus at all; if
he lies behind the “ancient legion” in book II, he does so as part of the
common background of epitomized sources. The evidence of Vegetius’
late-antique reconstruction of the “ancient legion” weighs against direct
use of so detailed and professional an authority as Patemus.'

John the Lydian cited Patemus as a Roman military rhetorician along
with Celsus, a mysterious Catilina, Cato and Frontinus, and Vegetius
himself.^ There is no reason to think that John the Lydian in the sixth
century, any more than John of Salisbury in the twelfth centurysaw
all of these works, though the Lydian clearly had access to Patemus,
Catilina is a shadowy figure, but most likely he was a later, third- or
fourth-century, epitomator of the authors in the same tradition as that
followed by Vegetius, and so a possible unnamed source used by him.

The study of Vegetius’ repetitions reveals his method best, as a
compiler of material epitomized from a very few secondary sources
which he then wrote up with literary embellishments in a classicizing
style according to a pre-arranged plan.^ He exercised reasonable care,
but not academic exactitude, throwing the material together quickly.

'Schenk, 23 ff.

^Epit. 11.7 itself seems to be a post-Constantinian list of officers and under-officers
inserted for supplementary purposes; V. had some but not enough ancient material for his
ambitious scheme to reconstruct the “ancient legion”. loh. Lydus de Mag. i.46 is a much
longer list derived from a related source.

Moh. Lydus de Mag. 1.47: iidpTupct; ICeXaot; te koi ndTspvo^ Kai KaiiXi vaq, ou/
6 auvcopoTTiq aXX' ^tepoc;, Kdicov <t8> Trpd autcov 6 rtpcotoc; Kai Opovif vog, pe0’
oOq Kai ‘Pevdioq, ‘Pcopaioi Tidvxec;. “as witnessed by Celsus, Patemus and Catilina, not
the conspirator but another, Cato before them, the first (writer), Frontinus, and after them
Renatus, ail Romans.”

^loh. Saresb., Policrat. (ad 1159) 6.19. (618a); adeat Catonem Censorium, legat et ilia
quae Cornelius Celsus, quae /ulius Iginus, quae Vegetius Renatus, cuius, eo quod
elegantissime et diligentissime rei m Hi tar is artem trad id it licet exempla perstrinxerit, plura
inserui. “Let him consult Cato the Censor, let him read also Cornelius Celsus, Julius
Hyginus, and Vegetius Renatus, of whom i have included much for the reason that he
treated the art of war most elegantly and thoroughly although he made limited use of
examples.” Cf. M. Manitius, Geschichte der lateinischen Literatur des Mittelalters III
(1931) 258: John excerpted Vegetius and Frontinus’ Strategemata.

^F. Lammert, Klio xxxiii (1940) 286, observed a number of repetitions in the chapters on
siege warfare, viz., Epit. IV.6 and 21, 5 and 24, 19 and 30, 22 and 29. These are due in
part to coverage of the same tactic from the point of view of the defenders and the
attackers of the city. There is a certain shift in treatment, as IV. 1-11 cover fortifications
and provisioning for a siege, whereas IV. 12-30 are on tactics of siege-warfare, in which
the presentation of attack and defence tactics is partly integrated, partly separate, but in
the main weighted towards defence, Lammert, Klio xxxi (1938) 399, was wrong to
schematize IV.8-11/12 as defence, IV. 12-30 as attack.

xxvi VEGETIUS’ EPITOME OF MILITARY SCIENCE

executing an intelligent general scheme. His work stands favourable
comparison with that of other contemporary semi-official authors at
Court such as Eutropius and Victor, who like himself combined the
writing of literature with the public life of busy hommes d’affaires.

General principles are obviously repeated because of their
importance.” Vegetius’ chapter on “General Rules of War” (III.26) is an
extended list of such principles, the majority of them repeated from all
parts of book III, but also including one or two from the first two
books.^ These rules were intended to provide an aide-memoire of the
main principles of field strategy and tactics; such recapitulation was a
valued technique of late antique didactic writers.^

The repetition of wider subjects'” is also attributable to the editorship
of Vegetius, as in the course of writing he came to use the same
material under different rubrics.^ His programme of reform necessarily

'e.g., Epit. III.3: saepius enim penuria quam pugna consumit exercitum, et ferro saevior
fames est^ “for armies are more often destroyed by starvation than battle, and hunger is
more savage than the sword”; Epit. 111.9: nam fames, ut dicitur, intrinsecus pugnat et
vine it saepius sine ferro., “for hunger, they say, fights from within, and often conquers
without a blow”; Epit. III.26: qui frumentum necessariaque non praeparat, vincitur sine
ferro., “he who does not prepare grain-supplies and provisions is conquered without a
blow.” Cf Front. Strat. 4.7.1: C. Caesar dicebat idem sibi esse consilium adversus
hostem, quod pie risque medic is contra vitia corporum, fame potius quam ferro superandi,
“C. Caesar used to say he had the same idea for the enemy, as many doctors have against
diseases of the body, that of overcoming them by starvation rather than the use of steel.”

^Epit. 111.26 p. 122.3 (Lang): Amplius iuvat virtus quam multitudo, “Bravery is of more
value than numbers.” Cf 1.8: In omni enim conflictu non tarn prodest multitudo quam
virtus, “For in any conflict it is not so much numbers as bravery that pays off’ (the
clearest example). A few are not repeated from any part of the text as we have it. The
most significant part of the list appears to be a summary of the seven “general
actions”— depugnationes — of III.20.

^Cf the list of necessariae sententiae in the probably contemporary work Palladius de
Agric. 1.6. The Byzantine military writer Maurice, at the end of the sixth century ad,
included in a similar list some Greek translations of V.’s rules; e.g., Epit. III.26 p.121.1-2
(Lang): In bello qui plus in agrariis vigilaverit, plus in exercendo milite laboraverit, minus
periculum sustinebit, “In war, he who spends more time watching in outposts and puts
more effort into training soldiers, will be less subject to danger.” = Maur. Strat. 8.2.2: ‘O
TiXeov CTUvotypuTivQ) V CTxpaTEUfiaTi mi nkcov tc^ yupvd^eiv touc; axpaxicoxac;
TTOvcDv eXdxicTxa »civ6uv8uei mxd xov TioXEpov.

^Epit. 1.9-20, 1.26-27, 11.23, III.4, III.9, III.IO (training); 1.21-25, III.8, HI.10
(castrametation); 1.20, U.15-17, 111.14 (battle-array); 1.4, 1.15-16, 1.17(7), 1.18(7), 1.20,
II.2, 11.15, 11.17, 11.23, III.14, III.16-18, 111.20, III.22, IIL24 (legionary light-armed);
11.22, III.5 (signals); 111.3, IV.7 (food supplies and civilian population); 11.10, III.2, IV.7
(care of sick); 1.3, I.IO, 11.24, III.4, 111.7 (swimming).

^Epit. I praef: per quosdam gradus et titulos antiquam consuetudinem conamur ostendere,
“we attempt to show, by a number of stages and headings, the ancient system...” A.
Andersson, Studia Vegetiana, ch. 3, argues convincingly on grounds of style that the
rubrics are by V. F. Lammert, rev. in Philologische Wochenschrift Ix (1940) 79, reminds

INTRODUCTION

XXVll

required him somehow to adapt revered ancient example to
contemporary conditions. This was a difficult task, and it led him to
treat of a subject from both the more theoretical and practical points of
view. Thus at 111,14-18 the legionary array is presented in a more
detailed formulation of that recommended already in 11.15-17, in line
with Vegetius’ perception of its practical application to field
campaigning, the topic of book III. Because of his chosen interest in
matters of internal legionary organization and structure in book II, on
the other hand, his earlier treatment serves a more theoretical aim.

Many repetitions are there to simply inform the author’s abundant
style. This seems the natural interpretation of repetitions of material in
close succession, e.g. in IV.26-28.' Such abundantia may also explain
repetitions placed farther apart, such as IV.22 and 1V.29, and 11.23,
summarizing (with slight variations due to Vegetius’ inaccuracy) the
chapters on training 1.9-19, 26-27. Likewise cases of the same exempla
occurring twice.^

The only section of the entire Epitome which does not exhibit
repetitions is that on shipbuilding and navigation (IV.33-42). Here he
will have had sufficient rather than too little material, as he implies at
one point.^ None the less, he did not hesitate to embellish it with
contemporary references in his usual style. Thus, Christian allusions in
IV.34 and IV.35, a barbarian word for British spy-boats, with probably
the whole exemplum concerning the spy-boats,'^ a circumspect reference
to the festival of the navigium Isidis and an official decision about the
date of Easter,^ and an allusion to Vergil’s Georgies 1V.41. The other
naval chapters IV.31-32, 43-46 exhibit all the familiar hallmarks of
Vegetius’ own editorship intended to advance his reforms, adorn his

us that such a framework is typical of tactical treatises such as Front!nus’ Strategemata.
Cf. also Aelian Tact, praef. 7. How far the plan is V.’s own, we cannot tell for sure; but
much of it subserves his reformist and polemical intentions, and is thus far likely to be
original.

‘Cf. also within the same chapter, IV.25 p. 146.1-8 (Lang), ibid. p.146.8-13, III.21
p.l 11.8-11, ibid, p.l 11.20-22. See P. de Jonge, “Ammianus and Vegetius”, in id. (ed.),
Ut Pictura Poesis: Studia latina Petro lohanni Enk septuagenario oblata (Leiden 1955)
104-105.

^Epit. 1.3 and 1.10, 11.11 and 1V.24, IV praef. and IV.26, cf IV.9.

^Epit. IV.40.

^Epit. IV.37, to be set beside the fact that V, is one of the earliest authors to use the
probably Celtic word dningus.

^See Epit. IV.39 note.

xxviii VEGETIUS’ EPITOME OF MILITARY SCIENCE

style and amplify his source-material, including the use of repetition,'
This pattern points to use of the same predigested sources as for the
bulk of the Epitome, sources which he revised, redesigned and rewrote
with a message for a contemporary audience, above all those in
government.

The accusation that Vegetius confused the institutions of diverse
periods of the Roman Empire is, therefore, beside the point. He was not
interested in telling the history of the Roman army; he was not an
historian, but something more akin to a politician, seeking to reform
contemporary institutions and strategic thinking. He was well aware that
the basic model was from Cato, albeit mediated through later authors
such as Celsus, Frontinus and Patemus, and to himself probably by later
epitomes of them. He largely supplemented gaps in the records available
to him by intelligent conjecture based on etymologies of old military
titles and knowledge of contemporary military institutions. He was not
trying to avoid confusion of historical institutions but to flesh out the
legion of Cato in a manner that would have relevance to the modem
field army. As a strategist, he presents the “ancient legion” throughout
as a model for the present. Ancient institutions and titles are set side-by-
side with modem. Lessons for contemporary practice are sought and
suggested, A truly historical approach was strictly irrelevant to his aims,
and lack of it should not be held against him.

The Epitome is, then, not a true Art of War, but a political and
strategic tract, an originally antiquarian account of Cato’s army tricked
out and rearranged as a commentary on present-day inadequacies. The
circumstances of composition reflect this result. Book I, we learn from
the preface to book II, was written first separately, off the writer’s own
bat as a paper for the Emperor’s instruction and benefit on the

'Thus, the list of provinces originally served by the two fleets at Misenum and Ravenna,
composed after the withdrawal of Aegyptus from Oriens in the late ad 360s {Epit IV.31,
cf 1.28, an editorial chapter; both lists will be of V.’s own composition), the deterrence-
theory of military preparedness restated at IV.31 in a naval context in terms derived from
III praef., the structure of the two naval legions under praefecti, and ten tribunes
commanding a cohort each, resembling V.’s reconstruction of the legio antiqua (IV,32,
cf. 11.12), the heavy arms of marines (IV.44, cf. 1.20 and IMS), the advantage of height
presented by turrets on the ships and the use of fire-darts described in terms which derive
from earlier comments on siege-warfare (IV.44, cf. IV.8, 17, 18, 19; IV.44, cf. IV. 18),
opportunities for ambush apparently deriving from those already listed for land warfare
(IV.45), finally, three dictionary definitions given for the asset, falx and bipennis (IV.46),
which could well have come from a reference book in V.’s library, cf. definitions of alae,
exercitus and legio (II. 1).

INTRODUCTION

XXIX

recruitment and training of Roman citizens, not barbarian mercenaries.
This was well-received and further examples from Antiquity were
commissioned by the Emperor—hence books II to IV, One Emperor
who we do know admired the duces of the Latin age (early and middle
Republic, 509-c. 100 BC) was Theodosius I (reigned ad 379-395),^ and
it is to him that the work is most likely directed. The work covers only
such areas of strategic and administrative practice as the author
considered in need of reform. The impression that it is an Art of War
derives from the fact that it covers much of the same ground and uses
basic material from such sources. But at best it gives only an incomplete
idea of the various departments of warfare.

What the Emperor really thought about it, as opposed to what he
apparently said, is anyone’s guess. But whereas all Emperors in the
possible period of writing from AD 383 to 450 made sweeping use of
barbarian mercenaries, it is evident that the driving-force behind
Vegetius’ reforms was the desire to reduce and down-grade the role of
all non-Roman ethnic forces. This is most demonstrable for book I
which is explicitly about the recruitment and training of legions of
Roman citizens, but it is also behind the “ancient legion” of book II and
the strategies for the combined infantry-and-cavalry field-armies of book
III with their heavy and light divisions all patterned after the legion of
Cato. Doubts are cast at the same time on the reliability and training of
barbarian auxilia (palatina) and, although the elite barbarian cavalry are
praised in 1.20, the discerning will have noticed that the “ancient legion”
in book II contained large numbers of “fully integrated” legionary
cavalry^ able to carry out the tactical functions outlined in 111.16 ff.
Vegetius’ insistence that the army could no longer build camps may also
be linked to the enrolment of large numbers of barbarians under their
own officers, who lacked technical knowledge of Roman military
traditions.^

Finally the whole work is informed by the need to carry on the fight
with the barbarians, both within and without the system, by land and by

‘Claud, pan. de Hon. cons. IV 399 ff, cf Ps.-Victor Epit. de Caess. 48.11-12.
^inflated to 726 by Vegetius, more than double the 300 normal in the age of Cato.

^So R. Grosse, “Das rOmisch-byzantinische Marschlager vom 4.-10. Jahrhundert”, Byz.Z.
xxii (1913) 95—96. Cf. Amm. 18.2.6: auxilianimilitessemper muniaspernenteshuiusmodi
ad obsequendi sedulitatem luliani blanditiis deflexi... “the auxiliary soldiers, who always
refuse labours of this kind, were persuaded to complete obedience by the blandishments
of Julian...” But contrast Amm. 31.8.9. For Ammianus’ disapproval of barbarians under
their own officers, cf. Amm. 31.16.8.

XXX

VEGETIUS’ EPITOME OF MILITARY SCIENCE

sea. For although Vegetius was aware that the barbarians had not yet
posed a naval threat, he was wise enough to foresee that eventuality
which was realized in the second decade of the fifth century AD.‘
Against a background of defensive operations, holding forts and cities
against invasion, controlling access to foodstuffs, resisting the enemy by
guerilla tactics and starvation, Vegetius proposed to purge the barbarian
preponderance within the army. Nor was he alone. Synesius called for
the purging of barbarians from the eastern field armies in
Constantinople, AD 399,^ and this was carried out by the party of the
praetorian prefect Aurelianus for whom he spoke, for we find Alaric
moves out of Dacia and Macedonia where he had been magister militum
per Illyricum from 397 and takes to invading Italy in 401 Earlier there
had been a general massacre of Gothic soldiers serving in the eastern
army after the disaster at Adrianople, AD 378, which was widely
regarded with satisfaction as a timely safeguard.'^

In the west the downfall of Honorius’ magister utriusque militiae
Stilicho in AD 408 was followed by a massacre of the families of Gothic
Roman soldiers, 30,000 of whom responded by deserting to join forces
with Alaric and precipitated the sieges of Rome in 408^09, and the
sack of AD 410.^ In AD 471 the Emperor Leo carried out another purge
of selected groups of Gothic and Alan soldiers in the east.^ The history
of the period shows the Roman governments reacting violently against
Gothic and other barbarian mercenaries after long periods of depending
on them, whenever the balance of power swung in their favour or they
could play off a new group of barbarians against those who had
outstayed their welcome or abused their paymasters. If Vegetius’ wish

'J.R. Moss, “The effects of the policies of Afitius on the history of western Europe”,
Historia xxii (1973) 711-731, esp. 723-728.

^Synes. De Regno (19) 23B = 1092C-1093A: f^6r| dvaKiriT^ov fipiv xd ‘Propaicov
4>povf||iaTa, Kai ouvEGiax^ov aOxoupyEiv xdq v(Ka<;, Koiv(Dv(a(; dvexop^vouq,
dXV dTia^iou vxac; dTcdaij xd^Ei t 6 pdppapov. “We must immediately recover the
courage of the Romans, and get used to winning our own victories, not putting up with
partners but dismissing the barbarian from every rank and post.”

^Cf. P.J. Heather, “The anti-Scythian tirade of Synesius’ De Regno", Phoenix xlii (1988)
152-172, id., Goths and Romans (Oxford 1991); T.D. Barnes, “Synesius in
Constantinople”, GRBS xxvii (1986) 93-112,

^Amm. 31.16.8, Zos. 4.26.

^Zos. 5,35,5-6.

'^’W. Goffart, “Zosimus, the first historian of Rome’s Fall”, American Historical Review
Ixxvi (1971) 429 = id. Rome's Fall and After {Londox\ 1989) 98 n.76 with refs., and P.J.
Heather, Goths and Romans (Oxford 1991).

INTRODUCTION

XXXI

to change the demography of the army was widely held, history suggests
that the late-Empire was in the grip of political and economic destinies
which made it a task of the utmost difficulty; such measures seem to
have succeeded, albeit partially and temporarily, only in the realm
controlled by the Byzantine government. The chances are that Vegetius’
Emperor, even if he agreed with his adviser in principle, was powerless
to carry out such a programme, at least in the western Empire,

Writing at the threshold of the Byzantine age, Vegetius was an author
for whom Cato and Frontinus were already “ancient” writers, as they are
for us. The past glories he set out to re-create played an inspirational
role for him, as they did for later potentates such as Justinian and
Charlemagne, and as we have seen for his own Emperor too. The spur
to his writing was provided by contemporary humiliations such as the
battle of Adrianople and its after-effects. The work offered a solution
based not on historical analysis, but the science and technique of
warfare as it had been practised by the Romans. Whatever the
shortcomings of the bald epitomes available to him, the author’s
judgement of his times was shrewd, and in some respects prescient, and
though a civilian his strategic thinking can frequently be confirmed by
parallels to contemporary professional soldiers.* Vegetius was thus a
significant late Roman writer who not only was well-informed about the
world of late Antiquity, but is in some ways a bridge beween the
mediaeval and classical Roman eras.

§2. The author

Nothing is known of Vegetius except what may be deduced from his
writings. Fortunately, we have two works to go on, the Epitome and the
Digesta Artis Mulomedicinae^ a veterinary work on horse and cattle
ailments, which tells us different things about the author which we
could scarcely have guessed. That it was the same Vegetius who wrote
both works was proved through close verbal and stylistic parallels by

'He was right in his assessment of the gravity of the threat to the survival of the Roman
Empire posed by the barbarian invasions, and correctly focused on the lack of available
professional field armies (after Adrianople) to meet it. He seems to have included naval
warfare in unstated anticipation of barbarians acquiring a naval capability, as later
happened. His belief that fortifications offered high security, both to soldiers and civilians,
was grounded in the military reality of the time, cf Amm. 31.6.4, 31.15-16. His emphatic
proposals to use food as a weapon and to harry the invaders by guerilla actions are
supported by strategies recorded and approved by Ammianus (noted in the commentary).

^E. Lommatsch (ed.), Vegetii Digesta Artis Mulomedicirtae (Teubner edn., 1903),

xxxii VEGETIUS’ EPITOME OF MILITARY SCIENCE

C. Schoener, and is generally accepted.’ The MS subscriptiones of the
Mulomedicina give Vegetius the praenomen “Publius”, as does the
oldest MS of the Epitome, the seventh-century “Excerptum Vaticanum”,
Vat, Reg. 2011? All other MSS of the Epitome call him Flavius
Vegetius Renatus. Since it was usual in the west to call imperial
servants “Flavius” plus their last name only,^ we should have expected
Flavius Renatus but this never occurs. It may be preferable to combine
the two strands of evidence and make him a true Flavius: thus Publius
Flavius Vegetius Renatus,"* However, true Flavii were rare. More likely
we have a reflexion of the practice (commoner in Constantinople)
whereby “Flavius” was treated unproblematically as an additional name,
adopted by imperial servants. Vegetius may have migrated to
Constantinople, like many Spanish hangers-on of Theodosius I, or else
the edition of his book prepared there by Eutropius may have affected
the form of his name. At any rate, he was probably named at birth
Publius Vegetius Renatus. It is also possible that he consciously chose
to use the dynastic title “Flavius”, to which he was entitled as an
imperial servant, in order for the proposals in the Epitome to have more
impact in Court circles, whereas he wrote the Mulomedicina for the
private amusement of himself and his aristocratic friends (see below)
who were too grand to be impressed by such a badge of service to the
regime.^

At a time when the old tria nomina system of Roman nomenclature
was all but dead,^ he was unusual in having a praenomen, a pattern
which recurs among only some fourth-century western senators and
some fourth- and even fifth-century curiales, town-councillors and their
families, from Italy and Africa.^ Although “Vegetius” is once attested

*C. Schoener, Studien zu Vegetius: Programm der kgl, bayer. Studienanstalt zu Erlangen
1887-1888 (Erlangen 1888) 18fr, W.S. Teuffel et al. (edd.), Geschichte der romischen
Literatur HI (Leipzig 1913^’) 317, M. Schanz, Geschichte der romischen Literatur IV
(Munich 1914') 198 f, RE Suppl. X (1965) s.v. Vegetius, col. 1018 (A.R. Neumann).

'Lang, xi.

^R.S. Bagnall, A. Cameron, S.R. Schwartz, K.A. Worp, Consuls of the Later Roman
Empire (Atlanta, Georgia 1987) 38.

"So PLRE L763, s.v. “Vegetius”.

^Cf. B. Salway, “A survey of Roman Onomastic Practice”, JRS Ixxxiv (1994) 124-145,
at 140.

‘^A. Cameron, “Polyonomy in the Late Roman Aristocracy: the Case of Petronius Probus”,
JRS Ixxxv (1985) 164-182, at 173, B. Salway, art. cit., 140-141.

'O. Salomies, Die romischen Vornamen. Studien zur romischen Namengebung. Soc. Sc.
Fennica, Comm. hum. litt. LXXXII (Helsinki 1987) 411-412.

INTRODUCTION xxxiii

as a signum, a name for domestic and familiar use only,’ there are
several second- and third-century examples of it as a gentilicium.^ It is
hazardous to draw firm conclusions from so small a number of cases,
but a Celtic or Celtiberian milieu for several holders of the name
emerges from the evidence, such as it is, and these Vegetii belonged to
the municipal gentry in Germania superior and a remote comer of
Spain. The process whereby a Latin name “Vegetus” borne by a
peregrinus in the western provinces was transformed into a gentilicium
“Vegetius” on the gaining of Roman citizenship was observed by
Schulze.^ There is a gap of several generations between these earlier
examples and our Vegetius, but it is plausible that he came of a long
line of municipal gentry owning estates in provincial Gaul or Spain,
who had risen to senatorial status.'’ The personal name or cognomen
“Renatus” is one of those which commonly denoted Christian beliefs.^
The Mulomedicina tells us further that he was a prominent breeder of
horses, had travelled the Empire extensively and was familiar with and

‘C/L 8.16561, Fortunatus Aug(usti) n(osth) adiutor a comm(entariis )... Vegeihi\ cf. 1608,
15626,15630, Tebessa, Africa Proconsularis, c. third century ad; I. Kajanto, Supernomina,
Soc. Sc. Fennica, Comm. hum. litt. XL.l (Helsinki 1966) 89.

^Rome: CIL 6.28400, A. Veg(etius) Diotrofes, Vegetia Victorina, second/third century.
CIL 6.1056 L2(42), Vegetius Firmus (1st coh. vigilum), AD 205. Picenum, at Septempeda:
CIL 9.5573, Vegetius Ingen(u)us (Roman army), second/third century. Tarraconensis, at
Conventus Bracaraugustanus: CIL 2.2381, Imp. Caes. T, Aelio Hadriano Antonino Aug.
Pioper T. Furnium Gal(eria) Proculum etA. Vegetium Gal(eria) Titianum, AD 138-161.
Narbonensis, atNcmausus: CIL 12.3826, L. Vegetius Ingenuus, prob. second/third century.
Germania superior, from nr. Borbitomagus: BRGK xxvii (1937) 73 no. 77, Mercurio
Augfusti) Vege[ti]us Gattus ob honorem aedilitatis posuit (aedile in Borbitomagus(?),
Celtic cognomen), late second/early third century. From nr. Eutingen-Niefem, Pforzheim:
BRGK xl (1959) 168 no. 124, Vegetius Patefrnus], Ve(getius) Severus, the former married
to Sulpicia Pattua (Celtic cognomen), second/third century. From Vitrey, Montigny-les-
Cherlieux, nr. Dijon: CIL 13.5910: Mercurio Saturninus Vegetii fi(lius) v. s. I m., prob.
second/third century. Raetia, nr. Castra Regina: CIL 3.5944: Veg(etius) Marcellin(us)
(Roman army), ad 222-235. The refs, are usefully collected by H. Solin, O. Salomies,
Repertorium nominum gentilium et cognominum Latinorum (1994).

^W. Schulze, Zur Geschichte lateinischer Eignennamen (Berlin 1933) 53: formed after
a copate cognomen and conceived as a patronymic, cf CIL 12.517, Sex Acutius Volt.
Aquila Acuto patri, beginning of first century AD.

^Consciousness of nobility was advertised by the use of the tria nomina in the late fourth
century, cf. Ausonius, Opuscula 16. Griphus ternarii numeri 80: tria nomina nobiliorum.

^I. Kajanto, The Latin Cognomina, Soc. Sc. Fennica, Comm. hum. litt. XXXVI.2
(Helsinki 1965) 135, 355.

xxxiv VEGETIUS’ EPITOME OF MILITARY SCIENCE

bred on his own studs many different breeds of horses.’ Significantly,
the breed for which he offers most information is the Hunnic warhorse.^
Goffart has drawn attention to the considerable knowledge of the
western barbarian world in Vegetius,^ which also includes incidentally
a number of Celtic and Germanic words. We may take it that Vegetius
was a man who knew his barbarians when he implicitly recommended
displacing them from the army.

We find from Symmachus’ attempts to hire race-horses for the games
staged in Rome for members of his family that Spain and Gaul were the
main centres of horse-breeding, carried on in estates presided over by
immensely wealthy senatorial grandees. Some of these senators may
have had literary interests. Symmachus calls one major supplier by the
signum or nickname “Euphrasius” which may allude to literary or
rhetorical skills."^ Vegetius also tells us that the demands of his friends
persuaded him to add book IV devoted to cattle—they were therefore
great landowners or possessores like himself^ It is Justified to place
Vegetius among their company, because the name though rare derives
from the cognomen “Vegetus” which is commonest in Spain, and next
commonest in Gallia Narbonensis.^

Also he evinces an eccentric and otherwise inexplicable interest in
Sertorius, the Roman nobleman who led an armed revolt from Roman
rule in Spain in the 70s BC.^ In the chaotic conditions of the late-
Empire, the central authorities in Italy were often unable to safeguard
the interests of territories on the fringes, so that provinces such as
Britain were repeatedly forced into taking measures for their own

^Mul. III.6.1: qui propter tarn diversas et longinquas peregrinationes equorum genera
untversa cognovimus et in nostris stabulis saepe nutrivimus. “I who am familiar with all
breeds of horses as a result of my travels which have been so various and far-flung, and
have frequently bred them on my own studs.” Mul. I prol. 6: cum ab initio aetatis
alendorum equorum studio flagrarem^ “since from my earliest years I was fired with zeal
for breeding horses.”

^MuL III.6.5.

^W. Goffart, Rome ’s Fall and After (1989) 69-70.

^Symm. Ep. 4.58-63.

^Mul. IV prol. 1-2.

*41 out of 77 examples in A, M6csy et al., Nomenclator, Diss. Pannonicae 3.1 (Budapest
1983). When his figures are adjusted to take account of the different sizes of the samples
in each western province, Spain still comes first with 34%, Gallia Narbonensis second
with 16%, Noricum, Gallia Belgica and the two Germanies, and Britannia roughly Joint
third with around 10-12%. Thus a “Vegetus” was twice as likely to come from Spain as
from Narbonensis.

’£p/7. 1.7, 1.9.

INTRODUCTION

XXXV

security at the price of revolting from the Emperor and putting up
rivals. The revolt of the Spaniard Maximus staged from Britain in AD
383 was made easier precisely by the excessive favour shown towards
Alan mercenaries by the Emperor Gratian in Milan at the expense of the
regular Roman troops.* Gratian is strongly criticized in 1.20 for military
decline in his reign and after.

From subscriptiones to the Epitome we learn that Vegetius was a v/>
inlusths and Count (First class), a rank reserved for the highest echelons
of the imperial bureaucracy or the army chiefs of staff. Since Vegetius
disclaims all personal military knowledge and only claims to compile
information from books for the Emperor or his generals to apply (I
praef., 1.8, 1.28, II praef., II.3, 11.4, 11.18, III.6 init., III.10 fm., 111.20
init.), we may take it for granted that he was a bureaucrat. There is no
particular internal evidence weighing in favour of Praetorian Prefect,
Count of the Sacred Largesses,^ or Count of the Privy Purse, although
his demonstrable interest in taxation and money-matters would suit all
three. But he might equally well have been Master of the Offices,
Prefect of the City or Quaestor of the Sacred Palace. Perhaps the most
satisfying suggestion is by Goffart that he was Count of the Sacred
Stable, frequently a vir inlustris from the early fifth century, at least.^
Vegetius’ extensive travels in connexion with horse-breeding, knowledge
of both warhorses and luxury civilian saddle-horses, and concern with
Hunnic and German breeds and the avoidance of fraud “practised upon
the country” could then be explained as arising from his official duties
in the procurement of military and civil mounts."* But one wonders that
Vegetius did not advertise the fact.

MS “n” has a subscriptio that he was comes sacrum (sic). It was
suggested by Schoener that sacrum is an abbreviation of sacrarum
(largitionum), but Vegetius’ references to money suit other high offices
as well as that of imperial finance minister. Rather, it may perhaps be
a corruption of comes stabuli. A mediaeval scribe would have been

'Ps.-Victor Epit. de Caess. 47.6, Zos. 4.35.2-6 - Eunapius fr. 51 (Blockley).

^“Sacred” in the later Roman Empire was a defining epithet of the imperial court and
institutions. For the various positions, see A.H.M. Jones, The Later Roman Empire
(Oxford 1964).

'W. Goffart, Traditio xxxiii (1977) 89-90 = id., Rome 's Fall and After (1989) 69-70. R.
Scharf, “Der comes sacri stabuli in der Spatantike”, Tyche v (1990) 135-147, esp. 145
and 137, assumes that V. was comes sacri stabuli under Valentinian III, c.440, without
argument.

Vu/. 1II.6.

xxxvi VEGETIUS’ EPITOME OF MILITARY SCIENCE

unlikely to understand comes stabuli as the title is rather rare, and the
corruption of an uncial stabuli to sacrum is not impossible, particularly
if the “ta” were transposed. The transition from “b” to “r” is easy, and
the “li” could be conRised with the final upright of “u” so as to suggest
three uprights and hence “m”. All that would remain is for uncial “t” to
be confused with “c”, which is also easy, since the downstroke of “t”
curled to the right. The result, though a false title, might seem more
satisfying to monks used to allusions to sacred things; sacrum was
presumably misinterpreted as an abbreviated neuter plural. Another
possibility is that the title was originally given in the form comes sacri
stabuli^ and the last element somehow dropped out; the corruption then
of sacri to sacrum is not unduly difficult.

Vegetius makes his Christian allegiances very clear in the prefaces to
both works, and the theocratic nature of the late-Roman state is affirmed
by the frequent collocation of God and Emperor in almost-equal
partnership. In Epit. II.5 he urges that soldiers should swear the oath of
allegiance by the Holy Trinity and the Emperor’s Majesty on the
Jesuitical grounds that they serve God by serving the Emperor, who is
God’s Representative. For the rest his analysis of Rome’s military
success is couched in wholly secular, pragmatic terms, and unlike many
contemporary bishops he shows no signs of belief in divine intervention,
or the orthodox thinking that Christian piety was a significant aid in
fighting or even itself a weapon against the barbarians.'

Alongside his Christianity he was a traditional admirer of Vergil’s
poetry and the superior wisdom which was attributed rather
superstitiously to him in the Latin-speaking half of the Empire, and an
enthusiast for Sallust’s histories and historical monographs; it is not
proven that he read Livy, in spite of much parallel material. His grasp
of Greek appears not to have been profound, since he abjures Greek
tactical authors in Epit, 1.8 and obtains Greek material in Latin versions
according to III praef and the prologue to MuL I, where he criticizes
the Latin style of Chiron and Apsyrtus, authors whose work ought to
have been in Greek.^ He seems in this typical of the late-fourth-century

‘Contrast Ambr. de Obitu Theod. 7, de Fide 2.136-140, Epp. 17.1; Maximus of Turin
{CCSL XXIII), Sermo 69.1-2, 83.1, 85.2.

^E. Lommatsch, op. cit., xxxvi-xxxvii.

INTRODUCTION

XXXVll

orthodox senatorial aristocracy in the west, steeped in the Latin classics
such as Vergil and Sallust, but on the whole ignorant of Greek.'

Finally he is consistently interested in offering explanations of
military customs which presuppose a medical point of view, is fond of
using medical metaphors— remedium^ medicina, vulnus —and touches on
medical concerns such as whether doctors or exercise are more
conducive to health, how cholera is spread, how to protect soldiers from
some of the dangers to health incidental to their profession, and the need
for pure water and keeping chickens for food for the sick. Such interests
are not too surprising in the author of the Mulomedicina, but one
wonders if Vegetius ever had any medical training or was purely self-
taught,

§3. The date

The upper limit is the death of Gratian, AD 383, for he is called divus
Gratianus at 1.20, and the lower AD 450, when an editor of the text
called Flavius Eutropius helpfully signed a subscriptio with details of
the place (Constantinople) and consular date, which has been copied in
the MSS of class e. Within these parameters debate continues. There is
a consensus, at least, that the work was produced in the western Empire
because of the mention of Gratian. Views on the identity of the
dedicatee mainly divide between Theodosius 1 (379-395) and
Valentinian III (425^55), though Honorius (393-423) has recently been
proposed as a possibility. The old view that it was dedicated to
Valentinian II (375-392), an adolescent cypher who never ruled in his
own name, rested on subscriptiones added by a late copyist.^ In what
follows I shall exclude personal arguments based on the use of imperial
titles, of which the evidence is too patchy to support firm arguments in
favour of this or that Emperor, and Vegetius’ rhetorical topoi offering
compliments to the Emperor, of which we have no fail-safe way of
judging the truth.^

*A,H.M. Jones, The Later Roman Empire (Oxford 1964) 11.987, P. Courcelle, Late Latin
Writers and their Greek Sources {Cdmhr\6%c^Mzs^. 1969) esp. 15 ff. (Symmachus), 20 ff.
(Macrobius), 78 ff. (Jerome), and 165 ff. (Augustine).

^Lang p.vii reports that the subscriptions to MSS D, fl, and V, three of the chief
representatives of his class x, are inscribed ad Theodosium imperatorem, and name him
elsewhere in the text too. These will be interpolations if Lang’s stemma is correct, but
T.D. Barnes, “The date of Vegetius”, Phoenix xxxiii (1979) 254-257 raises the suggestion
that class ti diverged from e even before Eutropius’ recension of ad 450.

Tor a fuller treatment of the date see Milner, Vegetius, ch, 3.

xxxviii VEGETIUS’ EPITOME OF MILITARY SCIENCE

Goffart and Birley have most recently joined the ranks of Seeck and
Gibbon in preferring Valentinian III.* Proponents of this reign object
that Gratian could not have been criticized to Valentinian II, his half-
brother, or Theodosius I, his half-sister Galla’s husband from AD 388,
five years after his death. This of course is debatable, but Theodosius,
at least, was considerably less closely related. Goffart argues that
Vegetius’ example of the Second Punic War (1.28) presupposes a
minimum time-scale of about two decades—the length of time from the
First to the Second Punic War—between the end of the “long peace”
and the series of disasters suffered at the hands of the Goths described
in 1.20. If the “long peace” were understood to be the reign of Gratian,
twenty years from his death would stretch to the early 400s when a
second Hannibal in the shape of Alaric was causing havoc in Italy,
which culminated in the sack of Rome in AD 410. But a reaction could
not set in until Honorius was dead and Valentinian III, untainted by the
disasters, could preside over a recovery engineered by his patricius
AStius; so it was this revival that Vegetius was in effect publicizing.^
Birley cites a number of “Novels” or decrees of Valentinian Ill which
allegedly attest such a military revival.^

However, Goffart pushes his timescale argument too hard. There is
no reason to separate the analogy of the twenty or so years of peace
between the First and Second Punic Wars from the “long peace” itself
identified with Gratian’s reign; moreover the figure “twenty” is not
relevant. The period of disasters is more naturally seen as including the
large-scale wars that must have terminated the “long peace” if the
phrase was to have any meaning at all—thus the disastrous campaigns
of AD 377-378 ending in defeat at Adrianople and the destruction of
two-thirds of the eastern field army along with the eastern Emperor
Valens, while Gratian was still on his way to join forces with him. The
sacking of cities can be explained by the Gothic incursion into

’E. Birley, “The dating of Vegetius and the Historia Augusta"', Bonner Historia Augusta
Colloqium 1982-83 (1985) 57-67 == id.. The Roman Army, Papers 1929-1986, Mavors
IV (Amsterdam 1988) 58-68; W. Goffart, art.cit.; O. Seeck, “Die Zeit des Vegetius”,
Hermes xi (1876) 61-83; E. Gibbon, Decline and Fall of the Roman Empire, ed. J.B.
Bury, Ill (1897) 187 n.l28.

^Goffart, 61 ff.

’Note that the chief witness, which Birley quotes at second hand from L. Virady, “New
evidences on some problems of the late Roman military organization”, A.Ant.Hung, ix
(1961), 333-396, is actually a Novel of Theodosius II, no. 24, from Constantinople, ad
443.

INTRODUCTION

XXXIX

Pannonia, Dacia, Thrace and Macedonia from 378 to the peace
settlements with Gratian and Theodosius I in the early 380s.' The
mention of Gratian was only worthwhile if he was still remembered,
which would not have been the case a generation later.

Secondly there was no military revival under Valentinian III. The
Novels merely show the State having increasing difficulties raising
recruits and money from the traditional sources of supply, so that it had
to resort to new-fangled taxes and the removal of ancient privileges—a
litany of decline, in fact. And whereas Vegetius aimed to restore a fully-
trained national Roman standing army, Valentinian Ill’s temporary
military successes were entire due to Aldus’ ability to hire Hun
mercenaries, a policy which would have been anathema to Vegetius.

C. Giuffrida argues for Honorius on different grounds extrapolated
from the allegedly pacific, philo-barbarian policy followed by
Theodosius I while he was alive and by Stilicho under Honorius until
his fall in AD 408, contrasted with the violently anti-barbarian senatorial
policy followed by the government of Honorius between 408 and the
sack of Rome in 410.^ She makes Vegetius propagandist for the
nationalist senatorial interest headed by the magister officiorum
Olympius, Her Vegetius would have been one of those directly
responsible for that untimely intransigence which hastened the demise
of Rome.

The argument for Honorius has the merit of explaining the mention
of Gratian since the blame could scarcely be pinned on Theodosius I,
the Emperor’s own father. It is also probable that the desertion of
30,000 Gothic soldiers upon the massacre of their families by the
Romans in Italy in 408 left them with the rump of a national army,
deprived of much of its barbarian stiffening. However the policy of
Theodosius I and Stilicho was hardly “philo-barbarian”, but involved
playing off one group of barbarians (e.g., the Tervingi) against another
(e.g., the Greuthungi, heavily defeated in AD 386),^ taking in some and
resisting others. Their opposition in principle to Vegetius’ plans should
not be assumed. Also we must not forget the protracted circumstances
of composition of the Epitome, book I of which appeared first and in a

'Pac. Pan. Lat. 2(12) 32.3-4, cf. Them. Or. 34.24.62-64 (ad 382), Hieron. Epp. 60.16.2
{CSEL LIV.M), c. AD 397, looking back over 20 years of violence.

^C. Giuffrida, “Per una datazione ^QWEpitoma rei militaris di Vegezio: politica e
propaganda nell et^ di Onorio”, Sic.Gymn. xxxiv (1981) 25-56.

^Hydatius Chron. ed. A. Tranoy (1974) a. 386. 13a.viii.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

separate edition from books II-IV, and in which the background of
military disasters rather tails off in the latter books instead of getting
worse, as we should expect.

In fact, the arguments for Theodosius I being the dedicatee are
stronger than for the other candidates. ‘ First, the period is right.
Vegetius alludes to the battle of Adrianople (III.l 1)^ but not to the sack
of Rome. Rome is cited three times, no less, as the example of the
inviolate city (IV praef., IV.9, IV.26), whereas this would be in poor
taste, surely, after ad 410. Also one of the most dangerous groups of
barbarians starting from their invasion of Gaul on 31st Dec. AD 406, to
wit the Vandals, are not named by Vegetius, whereas the Goths, Huns
and Alans, the victors of Adrianople, are (1.20, III.26). Moreover, he
calls the Huns and Alans one nation (III.26), alluding to an event
noticed also by Ammianus Marcellinus;^ but they split up when the
Alans joined the Vandals in 406,"^ and the Huns remained settled in
Pannonia until ad 427.^ Further, there was no naval war with the
barbarians at the time of writing (IV.31), but the Vandals acquired a
naval capability by ad 419,^ and it was well-known that Alaric had tried
to take to the sea at Rhegium in 410, and Wallia in southern Spain in
416.'

Further, it was still normal for coloni to be recruited (1.7), whereas
this was banned after pressure from senatorial landowners in the early
fifth century.* Gladiatorial games called forth no sanctimonious
denunciation from Vegetius, who was normally careful in matters of
piety (LI 1), but Honorius temporarily closed the gladiators’ schools in
AD 399, and this type of entertainment was dying out after 410 even in
Rome.’ Ravenna was the site of the old Augustan eastern fleet (IV.31),
but Vegetius did not mention that it weis made the new capital of the

'Cf. T.D. Barnes, “The date of Vegetius”, Phoenix xxxiii (1979) 254-257.

^W. Goffart, 63 n.85, G. Sabbah, 142.

^Amm. 31.3.1.

'’E. Stein, Histoire du Bas-Empire, tr. J.-R. Palanque (Paris 1949-59) 551 n.l61
^Marcellinus Comes, Chron. Min. 11.76.

^’J.R. Moss, “The effects of the policies of Aetius on the history of western Europe”,
Historia xxii (1973) 723-728.

^Oros. 7.43.12.

*Jones, 619.

Reallexicon Jiir Antike und Christentum XI (1981) 27-28 s.v. Gladiator IV.

INTRODUCTION

xli

western Empire from AD 401-402, equipped with major new
installations.'

There are more personal arguments, too. The anti-Gratian polemic
(1.20) is not strong enough to be addressed to Maximus, his murderer,
but too offensive to be addressed to Valentinian II, his half-brother.
Theodosius I occupied a position conveniently in between these
emotional poles, not being too closely related or politically involved in
Gratian’s realm. Secondly, a reference to a decision concerning the
calculation of the date of Easter (IV.35) can be identified with one of
Theodosius I’s theological reforms of AD 387-388, in which Theophilus,
bishop of Alexandria, worked out a paschal calendar for 100 years based
on Theodosius’ consulship in AD 380.^ Thirdly, Vegetius’ praise of his
Emperor’s city-founding programme suits an eastern Emperor, since this
activity was markedly more vigorous in the east than the west, and in
fact an impressive number of foundations can be attributed to
Theodosius I.^ Moreover, Theodosius is the one Emperor whose amateur
interest in Republican Roman history is recorded,"* and who we can be
sure would have liked to read Vegetius’ work from an antiquarian point
of view.

How then to reconcile a western author with an eastern Emperor?
First, Theodosius was the senior Augustus from AD 383 to his death in
395.^ Secondly, he stayed at Milan for three years following his victory
over Maximus in AD 388; without having to travel to Constantinople or
Thessalonica, Vegetius would have had ample opportunity to address
such a work to him, even on the subject of the reform of the field
armies, not just those of the west, and to meet and perhaps serve the
Emperor in person. It is consistent with the hypothesis that the work
seems to widen in scope after the polemic against Gratian’s army in
book I.

'M. Redd^, Mare Nostrum (Rome 1986) 659-660.

'Sabbah, 145.

^Milner, Vegetius, ch. 3 (d).

^C!aud.pa«. de Hon. cons. IV 299 ff., Ps.-Victor Epit. de Caess. 48.11-12, cf. SWA Alex.
Sev. 16.3, possibly inspired by a contemporary ideal.

^Oros. 7.35.

xlii VEGETIUS’ EPITOME OF MILITARY SCIENCE

§4. The late-Roman army

The classic, comprehensive survey of the late-Roman army is ch. XVII
of A.H.M. Jones’ Later Roman Empire. D. Hoffmann’s Das
spdtromische Bewegungsheer und die Notitia Dignitatum is the major
work of scholarship since. The late-Roman army is very different from
the army of the Principate, but grew out of it under the reforms of
Constantine, who built on a duality set up by Diocletian’s field army
and frontier defences, but relied largely on ^lite barbarian mercenaries
to bolster his own position in the State. But Vegetius criticized it against
the standard of the army of the middle Republic as transmitted by
sources from the Principate. Hence the non-specialist reader requires an
up-to-date synopsis of all three armies, middle Republican, Principate
and late-Roman. He is respectfully referred to The Roman World, ed, J.
Wacher, with articles on “The army of the Republic”, by G.R. Watson,
“The imperial army”, by Alistair Scott Anderson, and “The army of the
late Empire”, by R.S.O. Tomlin.’

For an understanding of Vegetius, the last of these armies is perhaps
least familiar and most important. Briefly, the late-Roman army in the
late-fourth century AD was of two main kinds, (I.) regular and (II.)
mercenary. (I.) Regular units comprised three categories, (1.) infantry
and cavalry and river patrols distributed around the frontier provinces
as limitanei under the command of duces who controlled sectors called
“duchies”, (2.) ^lite units of infantry and cavalry called comitatenses,
concentrated in important cities and imperial capitals to provide field
armies which could be quickly mobilized to deal with any large-scale
invasion beyond the resources of the limitanei, and (3.) elite units of
mounted Imperial Guard called scholae palatinae, who protected the
Emperor’s person and fought with the comitatenses. Of the mobile or
“field” troops, the scholae palatinae were very heavily barbarized, as
were also the cavalry vexillationes and the infantry avail ia palatina
among the comitatenses, leaving only the legiones as less barbarized and
having more Romans in them. It is these field troops that Vegetius was
writing to reform, by creating units which combined the specializations
of the various comitatenses under the umbrella of reformed and greatly
enlarged legions, with their more Roman military traditions and
manpower.

'The Roman World, ed. J. Wacher (London-New York 1987), Vol. I, Pt 3, pp.75-135,
with further bibliography.

INTRODUCTION

xliii

(IT) The cause was lost, however. The use of mercenaries at precisely
the time of writing was sharply on the increase, at the expense of
keeping fewer units of regulars and failure to maintain paper-strengths
of existing units, a process particularly evident in the West. The reign
of Theodosius I was a watershed for the increased reliance on
mercenaries, many or most of whom were (1.) foederati, treaty-troops
raised from barbarians settled both within and without the Empire, while
others will have been (2.) buccellarii, private armies raised by individual
barbarian leaders and even Roman generals from motley sources,
although the distinction between these terms became increasingly
blurred. (3.) A third source of such irregulars was prisoners-of-war or
dediticii.^ The horde-type armies with which Theodosius engaged the
usurpers Maximus (388) and Eugenius (394) marks the start of a new
era in which Roman governments were chiefly dependent on barbarian
condottieri for temporarily hired military forces which they no longer
maintained themselves. The mobile legions which Vegetius tried to save
and revivify disappear from the historical record in the western Empire
in the next century, although many of their counterparts survived
unreconstructed in the East into the sixth century at least.

'See J.H.W.G. Liebeschuetz, Barbarians and Bishops Army, Church, and State in the
Age of Arcadius and Chrysostom (Oxford 1990) 32 ff, A.H.M. Jones, 663 tf.

Ancient Synopsis'

The First book discusses the selection of recruits,^ from what localities
what sort of men should be approved as soldiers, and in what military
exercises they should be trained. The Second book comprises the
organization of the ancient army, in accordance with which an infantry
army may be instituted.^ The Third book illustrates all kinds of strategic
skills which seem necessary to land warfare. The Fourth book lists all
the machines used to attack and defend cities; precepts of naval warfare
also are appended.

Book 1
Preface.

In ancient times it was the custom to commit to writing one’s studies
in the liberal arts and offer them summarized in books to Emperors."*
For nothing is begun rightly unless after God the Emperor favours it,
nor is it appropriate that anyone should have superior or wider
knowledge than the Emperor, whose learning can benefit all his
subjects. That Octavian Augustus^ and good Emperors after him readily
thought so is clear from many examples. Thus with the testimonials of
rulers, eloquence flourished, while boldness was not rebuked,^ Spurred
on to emulate this, when 1 reflected that Your Clemency was better able
than others to excuse temerity in literature, I hardly felt my great
inferiority to ancient authors. Yet in this opuscule neither linguistic
elegance nor intellectual acumen was needed, but painstaking and
faithful labour, to put into the public domain for the benefit of Rome

'There is no reason to think that the Synopsis was not written by V.

^Like the Theodosian Code, V. uses iunior and tiro synonymously for “recruit”.

’V. is mainly concerned to reverse a supposed decline in the Roman infantry armies
(legions) at the expense of mainly barbarian 6Hte cavalry {vexillationes or equites) and
light infantry {auxilia palatina) —cf. 1.20, II.I, 11.3, 111.9—but in fact large numbers of
legionary cavalry too are included in the “ancient legion” described in Book II.

^Aelian Tact. 1.7 cites Plato’s Leges 625e-626b on the nomothete of the Cretans for the
view that art of war is the most useful of all, since mankind is on a constant war-footing,
there being by nature undeclared war between all cities. Cf III praef. III. 10. Celsus’
encyclopaedia contained four “arts”—agriculture, medicine, rhetoric and military science.
Polyb. 9.20.9 called war “the most honourable and serious of all arts”.

^The first Emperor of Rome, reigned 27 BC-AD 14, never called “Octavian Augustus”
until by historians of late antiquity, cf Amm. 26.1.13, Viet. Caess. 1.1, Eutrop. Brev.
1.12.2, Rufius Festus Brev. 19.

^’Cf Verg. Georg. 1.40: da facilem cursum atque audacibus adnue coeptis. “Grant an easy
course and support my bold undertaking.”

VEGETIUS’ EPITOME OF MILITARY SCIENCE

matters which lay scattered and hidden in the pages of various historians
and teachers of military science,'

We attempt to show then, by a number of stages and headings,^ the
ancient system of levying and training recruits. Not that those things
would appear unfamiliar to you, Invincible Emperor,^ but so that you
may recognize in your spontaneous dispositions for the safety of the
State the principles which the builders of the Roman Empire long ago
observed,'* and in this little book find whatever you think needful to
affairs of State, which are ever pressing.

(Recruitment, ch. 1-7)

1. That the Romans conquered all peoples solely because of their
military training.

In every battle it is not numbers and untaught bravery so much as skill
and training that generally produce the victory.^ For we see no other
explanation of the conquest of the world by the Roman People than
their drill-at-arms, camp-discipline and military expertise. How else
could small Roman forces have availed against hordes of Gauls? How
could small stature have ventured to confront Germanic tallness?^ That
the Spaniards surpassed our men not only in numbers but in physical
strength is obvious. To Africans’ treachery and money we were always

'The motif of inability to do justice to the subject was common in encomiastic literature,
cf E. Curtius, European Literature and the Latin Middle Ages, tr. W.R. Trask (New York
1953) 83 ff., L.B. Struthers, “The rhetorical structure of the encomia of Claudius
Claudian”, HSCPh xxx (1919) 57.

^This indicates that the rubrics are the author’s; their style is homogeneous with the text,
and they share in the same systems of variatio, cf A. Andersson, Studia Vegetiana
(Uppsala 1938) ch. Ill, pp.44^7.

^MS /) reads princeps invicte O Theodosi divorum augustorumpraecellentissime, probably
an interpolation. “Invincible Prince, O Theodosius, most Excellent of the deified Augusti.”

''i.e., during the middle Republic, third to second centuries BC, the age of Cato the Censor.
V. evades the problem how to inform the “omniscient” by attributing to the Emperor full
knowledge of what he is about to tell him anyway.

*This sentence belongs here in the “vulgate”; in the best MSS it is placed at the head of
the capitulation of Book I. The vulgate is to be preferred because the sentence does not
fit the capitulation and otherwise I.l would open with an inferential conjunction enim,
contrary to V.’s usage.

^’Cf Amm. 16.12.47 on Roman discipline being a match for German height and strength,
cf Caes. Bell. Gall. 2.30.4: plerumque omnibus Gallispraemagnitudine corporum suorum
brevitas nostra contemptui est, ibid. 4.1.9, 1.39.1. “Because of the excessive size of their
bodies our shortness has usually been a subject of contempt for all the Gauls.”

BOOK I: RECRUITMENT AND TRAINING

unequal.’ No one doubted that we were surpassed by the arts and
intelligence of the Greeks.^ But what succeeded against all of them was
careful selection of recruits, instruction in the rules, so to speak, of war,
toughening in daily exercises, prior acquaintance in field practice with
all possible eventualities in war and battle, and strict punishment of
cowardice.^ Scientific knowledge of warfare nurtures courage in battle.
No one is afraid to do what he is confident of having learned well. A
small force which is highly trained in the conflicts of war is more apt
to victory: a raw and untrained horde is always exposed to slaughter.

2. From what regions recruits should be levied.

The order of our subject demands that the first part should treat of the
provinces and peoples from which recruits should be levied. Now it is
common knowledge that cowards and brave men are bom in all places.
However, nation surpasses nation in warfare, and climate exerts an
enormous influence on the strength of minds and bodies. In this
connexion let us not omit what has won the approval of the most
learned men.^ They tell us that all peoples that are near the sun, being
parched by great heat, are more intelligent but have less blood, and
therefore lack steadiness and confidence to fight at close quarters,
because those who are conscious of having less blood are afraid of
wounds. On the other hand the peoples of the north, remote from the
sun’s heat, are less intelligent, but having a superabundance of blood are
readiest for wars. Recruits should therefore be raised from the more
temperate climes. The plenteousness of their blood supplies a contempt
for wounds and death, and intelligence cannot be lacking either which

'alluding to Jugurtha’s dictum (Sail. Jug. 35.10); urbem venalem et mature perituram, si
emptorem invenerit. “(Rome) ...a city for sale and soon to pass away, if it finds a buyer.”

^alluding to the likes of Verg. Aen. 6.847-853 and Horace’s dictum: Graecia captaferum
victorem cepit et artes / intulit agresti Latio, (Epist. 2.1.156-157). “Captive Greece
captured its fierce victor and brought the arts to wild Latium.” The whole argument
resembles Cic. de Harus. Resp. 9,19.

'Front. Strat 4.1-2, Val. Max. 2.7. Polyb. 6.37.9 ff,

'‘Small, highly-trained (legionary) armies are at the core of V.’s recommendations, cf II.4,
IIl.l.

^Arist. Pol, 1327b adapted by Posidonius, probably transmitted to V. through Varro, cf.
W, Theiler (ed.), Poseidonios, Die Fragmente (Berlin 1982) 11.72 fr, 71. The fullest
expression of it may be found in Vitr. 6.1. V, omits the climax that the perfect mix of
qualities was to be found in the Roman People and peoples of Italy, but allows the
inference to be drawn that barbarians were unsuitable.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

preserves discipline in camp and is of no little assistance with counsel
in battle.

3, Whether recruits from the country or from the city are more
useful.

The next question is to consider whether a recruit from the country or
from the city is more useful. On this subject I think it could never have
been doubted that the rural populace is better suited for arms.' They are
nurtured under the open sky in a life of work, enduring the sun, careless
of shade, unacquainted with bathhouses, ignorant of luxury, simple-
souled, content with a little, with limbs toughened to endure every kind
of toil, and for whom wielding iron, digging a fosse and carrying a
burden is what they are used to from the country.

Sometimes however necessity demands that city-dwellers also be
conscripted. These, when they have given in their names for military
service, must first learn to work, drill, carry a burden and endure heat
and dust; they must adopt a moderate, rural diet, and camp now under
the sky, now under tents.^ Only then should they be trained in the use
of arms and, if a long campaign is in prospect, they should be detained
for considerable periods on outpost-duty and be kept far away from the
attractions of the city, so that by this means their physical and mental
vigour may be increased.^

It is undeniable that after the City was founded'^ the Romans always
set out for war from town. But in those days they were not enervated
by luxury. Youth would wash off sweat collected in running and field
exercises, swimming in the Tiber.^ The same man was both warrior and
farmer, merely changing the style of equipment. This was so far true
that by all accounts Quinctius Cincinnatus was offered the dictatorship

'Cf Cato de Agric. praef. 4: at ex agricolis et viri fortissimi et milites strenuissimi
gignuntur. “From farmers both the bravest men and the strongest soldiers are bom.”

^Cf. Hor. C. 1.8.4: cur aprtcum \ oderit Campum patiens pulveris atque solis, “...why he
shuns the open Campus (Martius), having learned to endure dust and sun.” Onas. 10.5:
OaXireCTiv daKidaxoiq Kai KpupoCq umiOpou; peva. “...exercising

in summer heat without shade and in the icy cold in the open air.”

^On the mental as well as physical benefits of training, cf 11.23, Onas. 9.2-3.

^The traditional date for the foundation of Rome is 753 BC.

^Porph. ad Hor. C. 3.7.25, 1.8.8. V. repeats the example at 1.10. Cf also Cic. pro Cael.
36. A grassy plain in Rome along the Tiber, consecrated to Mars, the Campus Martius
was the place for army musters and exercises, and meetings of the comitia centuriata, the
Roman People in military order, during the Republic. From the first century bc onwards
it was gradually filled in with monumental buildings.

BOOK I: RECRUITMENT AND TRAINING

while he was ploughing.’ From the country, then, the main strength of
the army should be supplied. For, I am inclined to think, a man fears
death less if he has less acquaintance with luxury in his life.^

4. At what age recruits should be approved.^

Next let us examine at what age it is appropriate to levy soldiers. Indeed
if ancient custom is to be retained, everyone knows that those entering
puberty should be brought to the levy."^ For those things are taught not
only more quickly but even more completely which are learned from
boyhood. Secondly military alacrity, jumping and running should be
attempted before the body stiffens with age. For it is speed which, with
training, makes a brave warrior. Adolescents are the ones to recruit. Just
as Sallust says: “Directly as soon as youth was able to endure war, it
learned military practice in camp through labour.”^ For it is better that
a trained young man should complain that he has not yet reached
fighting age, than that he should regret that it has passed.^

He should also have the time to learn everything. For the art of war
does not seem a slight or trivial matter, whether you wish to train a
cavalryman, a foot-archer or a scutatusj or teach all the routines and all

'Traditional date, 458 BC. Cf. Livy 3.26.9, Eutrop. Brev. 1.17.1: is, cum in opere etarans
esset inventus... “while he was found at work, ploughing...” Livy also reports an
alternative tradition that he was digging a ditch.

^Alcibiades was said to have declared that the Spartans welcomed death in battle because
of the tough regime by which Sparta lived; cf. Aelian Var. Hist. 13.38.7.

^The scrutiny of recruits known as probatio or “approval” was carried out by the civil
authorities, often provincial governors in person, to check the identity, height, age and
status-qualifications of candidates for military service, before despatch to a unit. Cf R.W.
Davies, “Joining the Roman Army”, Service in the Roman Army, ch. I, 3-30.

'‘V. exaggerates; the ancient evidence was for recruitment from 16 years at the earliest,
cf Livy 22.57.9, Serv. ad Verg. Aen. 7.162. But from ad 364 Valentinian I allowed the
enrolment of young sons or relatives of established domestici in the 61ite scholae
protectorum domesticorum, which functioned like a staff college for military commanders;
the boys were to stay in sedibus —at the home base—but drew pay and rations of four
annonae. In time they would be old enough to bear arms and go on campaign; cf CTh
6,24.2-3, and Jones, 638. In the fourth century ad recruitment to ordinary army units
began at 19 years, cf CTh 7.13.1 (ad 326 S.), 7.22.4 (=12.1.35) (ad 343 S.).

^Sall. Cat. 7.4.

^Elderly soldiers were not uncommon in the late Empire, perhaps because they could be
enlisted at as late as 35 years, cf Jones, 616 n.l9, 635 n.60, J.R. Rea, “A Cavalryman’s
Career, ad 384 (?)-40r’, ZPE Ivi (1984) 84, letter iii. But cf Tac. Ann. 1.35.2, for men
who had served 30 years and more in AD 14.

’infantryman armed with a shield and darts, cf 1.17, 11.15, III. 14. V. also includes light
scutati (as well as heavy) at III.14, at least, who seem likely to be late-antique.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

the gestures of the armatura,' not to desert one’s post, not to
disorder the ranks, to hurl the javelin with a true aim and great force,
to know how to dig a fosse and plant stakes in scientific fashion, handle
a shield and deflect oncoming missiles with oblique movements, avoid
a blow intelligently and inflict one boldly. For this recruit so trained,
fighting against all manner of enemies in battle will be no terror but a
delight.

5. At what height recruits should be approved.

The height of recruits was, I know, always required to be up to the
incomma,^ so that men of 6 ft. (= 5 ft. 914 in., 1 -77 m.)^ or at least 5 ft.
10 in. (= 5 ft. 714 in., 1 ‘72 m.) were approved for the alares cavalry"* or
the First cohorts of the legions.^ But in those days the population was
greater, and more followed a military career. For civilian careers did not
then take away the better class of youth.^ So if necessity demands, it is
right to take account not so much of stature as of strength. Even Homer
himself is not wanting as a witness, since he records that Tydeus was
small in body but a strong warrior.^

6. That the potentially better recruits are recognized at selection
from the face and physical posture.

He who is charged with carrying out the levy procedure should take
great pains to choose those able to fill the part of soldiers from the face,
from the eyes, from the whole conformation of the limbs. For quality
is indicated not only in men, but even in horses and dogs, by many

'A special and perhaps ancient drill, practised as a sport by late-Roman officers and
Emperors, considered by V. to be the hallmark of the ancient legionary; cf. 1.13.

^Greek for “incised mark” - standard height.

^Roman measures (modem equivalents in brackets): the Roman foot measured 295 *7 mm.,
shrinking to 294*2 mm. in the third century ad, as opposed to the modern imperial foot
of 304*8 mm., cf. F. Hultsch, Griechische undromische Metrologie (Berlin 1882^) 94-97.
^i.e. service in the alae.

^CTh 7.13.3 (AD 367) reduced the minimum height for Italians for entry into unspecified
regular units from 5 ft. 10 in. to 5 ft. 7 in, (= 5 ft. 4% in., 1*64 m.); cf also 7,1.5
showing that one could be too weak or small for military service; 7.22.8, showing that
lower physical standards were applied for recruits for the ripenses than for the
comitatenses.

^'Cf. CTh 7.22.6-10 (ad 349-380), imperial constitutions forbidding veterans’ sons entry
to the civil service, apparently widi less than total success.

’Horn. II. 5.801.

BOOK I: RECRUITMENT AND TRAINING

points, as is understood in the teaching of the most learned men.’ Even
in bees, the Mantuan author says, it is to be observed;

“Two kinds there are, the better by its face
Distinguished and bright with ruddy scales;

The other type is shaggy and inert

And drags along its fat, cowardly paunch.”^

So let the adolescent who is to be selected for martial activity have
alert eyes, straight neck, broad chest, muscular shoulders, strong arms,
long fingers, let him be small in the stomach, slender in the buttocks,
and have calves and feet that are not swollen by surplus fat but firm
with hard muscle. When you see these points in a recruit, you need not
greatly regret the absence of tall stature. It is more useful that soldiers
be strong than big.^

7. Of what trades recruits should be selected or rejected.

The next matter is for us to examine from what crafts soldiers should
be selected or rejected utterlyFishermen, fowlers, pastrycooks,^
weavers and all who shall seem to have dealt in anything pertaining to
textile-mills^ should in my view be banned far from camp. Masons,
blacksmiths, wainwrights, butchers and stag- and boar-hunters may

‘Varro Rer. Rust. 2.7.4-5 (points of a horse), 2.9.3-4 (dog), 2.5.7-8 (cattle), Verg. Georg.
3.49-59 (cattle), 72-88 (horses).

^Verg. Georg. 4.92-94.

The selection of recruits according to lists of points useful in stockbreeding is not
attested in the sources, which speak merely of statura and robur, and may well be V.’s
own idea.

^Cf. CTh 7.13.8 (ad 380), forbidding slaves, innkeepers, brothel-workers, cooks, bakers,
or practitioners of “degrading” occupations from being recruited to the field army (?) or
at least the “cavalry”. The Romans had persistent and deep-seated qualms about the moral
qualities of various kinds of tradesmen, cf. Cic. de Off. 1.150. Those who pursued
dishonourable professions included infames such as actors, gladiators, charioteers and
brothel-keepers, cf. M. Kaser, “Infamia und ignominia in den rOmischen Rechtsquellen”,
ZRG Ixxiii (1956) 220-278. The list of professions that disqualified one from military
service, or at least the higher branches of it, was apparently considerably longer in the late
Empire, and included e.g. weavers and linen-workers, on which see below.

Purveyors of luxury foods were notoriously prized by officers for their tables; cf Ambr.
de Elia 46 on officers’ parties, and Pac. Pan. Lot. 2 (12) 14.3, SHA Claud. 14.11 on
fowlers, fishermen and huntsmen enrolled in the army for the purpose.

^'gynaecea = lit. “women’s quarters”, but denoting in the fourth century State-owned
textile factories, making among other things military uniforms. Weavers and textile
workers were among the many categories banned by law from the army. Such workers
were originally State-slaves, although by the mid-fourth century their condition had so far
improved that they had become de facto free persons bound by an hereditary tie to their
trade, cf Jones, 836.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

usefully be joined to the military. This is a matter on which the safety
of the entire State depends, that recruits be levied who are outstanding
both in physique and moral quality. The strength of the realm and the
foundation of the Roman Empire depend on the initial examination of
the levy.' Let it not be thought an unimportant duty, nor one which may
be delegated to anyone, anywhere. It is well-known that among his
considerable range of qualities Sertorius was praised by the ancient
(writers) for this in particular.^ For the youth in whose hands is to be
placed the defence of provinces, the fortune of battles, ought to be of
outstanding breeding if numbers suffice, and morals. Decent birth makes
a suitable soldier, while a sense of shame prevents flight and makes him
a victor.^

For what benefit is there in training a coward, of his spending several
years’ service in camp? An army never makes fast progress if the
selection procedure in approving recruits has been awry. And as we
know from practice and experience, it is from this cause that so many
defeats have been inflicted on us everywhere by our enemies, whilst
during long years of peace the levying of soldiers has been neglected,
while all those of decent birth have been pursuing civilian careers, while
recruits levied from landowners have through the corruption or neglect
of those granting approval been Joined to the army only when they were
of the sort their lords disdained to keep.'' So suitable recruits should be
levied with great care by great men.

'On physical strength and good character of recruits, cf Jos. Bell. lud. 2.580-582, 5.306,
cf. 6.38, Dig. 49.16.4.1-9 (Menander), 49.16.2.1 (Menander), 49.16.16 (Paulus) showing
that convicted felons were not allowed in the army. In normal circumstances freedmen
were not allowed to enlist, either.

^As quaestor in 90 BC Q. Sertorius early distinguished himself recruiting troops and
procuring arms for use in the Social War, cf. Plut. Sert. 4,1. Later, exiled from Rome, he
organized the Spanish tribes and equipped them with Roman-style arms to fight Rome on
equal terms from 80 BC until his assassination in 73 or 72, cf. Plut. Sert. 14, Flor. 2.10.

^On shame, cf. Ajax’s advice, Horn. II 15.561-564. On decent birth, cf Tiberius’
complaint, Tac. Ann. 4.4.4.

^Cf Amm. 19.11.7, 31.4.4, on the detrimental effect of using barbarian manpower to fill
positions in the regular army, merely because of the reluctance of Romans to serve. On
the recruitment of coloni or tenantry, resisted by senatorial landowners even during the
war with Gildo in ad 397, and subsequently banned in the early fifth century, see Jones,
184, 619; V. must have been writing before this ban. Cf Anon, de Reb. Bell. 4,
denouncing corrupt exactores from the governor’s officium involved in raising recruits,
and Symm. Ep. 9.10.2, denouncing the local authorities for the same.

BOOK I: RECRUITMENT AND TRAINING

(Training, ch. 8-28)

8. When recruits should be marked.'

The recruit should not be tattooed with the pin-pricks of the official
mark as soon as he has been selected, but first be thoroughly tested in
exercises so that it may be established whether he is truly fitted for so
much effort. Both mobility and strength are thought to be required of
him, and whether he is able to learn the discipline of arms, whether he
has the self-confidence of a soldier. For very many, though they seem
not unacceptable in appearance, are yet found unsuitable in training.
Therefore the less useful ones should be rejected and in their place the
most energetic should be substituted. For in any conflict it is not so
much numbers as bravery that pays off.^

So once the recruits have been tattooed the science of arms should be
shown them in daily training. But neglect due to long years of peace has
destroyed the tradition of this subject. Whom can you find able to teach
what he himself has not learned?^ We must therefore recover the ancient
custom from histories and (other) books. But they wrote only the
incidents and dramas of wars, leaving out as familiar what we are now
seeking. The Spartans, it is true, and the Athenians and other Greeks
published in books much material which they call tactica, but we ought
to be inquiring after the military science of the Roman People, who
extended their Empire from the smallest bounds almost to the regions
of the sun and the end of the earth itself This requirement made me
consult competent authorities and say most faithfully in this opuscule
what Cato the Censor wrote on the science of war, what Cornelius
Celsus, what Frontinus thought should be summarized, what Patemus,
a most zealous champion of military law, published in his books, and

‘C.P. Jones, “Stigma: tattooing and branding in Graeco-Roman antiquity”, JRS Ixxvii
(1987) 139-155, has established that soldiers were tattooed, not branded, without noticing

general principle of V. is the belief in the superiority of quality to quantity. On the
four or so months’ preliminary testing of recruits, not attested elsewhere, cf. 11.5. This
may be an invention of V.’s, or it may have been used by such units as regularly filled
vacancies from a large pool of soldiers’ sons or adcrescentes, where they would have been
able to pick and choose, cf. II.3, Anon, de Reb. Bell 5.7-8.

^Partly rhetorical hyperbole, partly true statement (so far as V. was concerned). V.
exaggerates his points for better effect. For the continued existence of at least some
contemporary training, cf. Ambr. de Off. 1.32, Synes. de Regno 13-14 (14 A-D), 19 (21
D).

VEGETIUS’ EPITOME OF MILITARY SCIENCE

what was decreed by the constitutions of Augustus, Trajan and Hadrian.*
For I claim no authority to myself, but merely write up the dispersed
material of those whom I have listed above, summarizing it as if to
form an orderly sequence.^

9. Recruits should be trained in the military step, in running and in
jumping.^

So, at the very start of the training recruits should be taught the military
step."* For nothing should be maintained more on the march or in battle,
than that all soldiers should keep ranks as they move. The only way that
this can be done is by learning through constant training to manoeuvre^
quickly and evenly. For a divided and disordered army experiences
danger from the enemy which is always most serious. So at the military
step 20 miles should be covered in five hours, at least in summer time.^
At the full step,’ which is faster, 24 miles should be covered in the same

'Perhaps Trajan is introduced merely to please Theodosius I, who liked to be thought of
as a second Trajan, for at 1.27 only Augustus and Hadrian are mentioned. Cf Ps.«Victor
Epit. de Caess. 48.8-10, K.H. Waters, “Trajan’s character in the literary tradition”, in
Polls and Imperium; Studies in honour of Edward Togo Salmon (Toronto 1974) 238-240,
Claud, pan de Hon. cons. IV 19, Them. Or. I4.205a. Also Pac. Pan. Lat. in editing such
a collection headed by Pliny’s and his own panegyrics may have been hinting at a
Trajan-Theodosius parallel.

^P, Taruttienus Patemus, ab epistulis Latinis to Marcus Aurelius in the AD 170s and then
praefectus praetorio almost certainly under Marcus Aurelius and Commodus, then
Commodus alone c. ad 180-82, wrote a juristic work on the Roman army of which a few
fragments survive, cf. Dig. 49.16.12, 50.6.7, loh. Lydus de Mag. 1,8, 47. This was
probably the source for the imperial constitutions, cited by V. at 1.27 only. Repeated signs
of late-Roman educated guesswork and reconstruction based on jejune material suggest
that V. had only late epitomes of Patemus and Frontinus, which subsumed the earlier
authors; thus, none of the named authorities was directly used. The second source-notice
at 11,3, which serves for the rest of the work, names Cato and Frontinus and “many
others”, i.e. the same collection of material as here. The account of the Ancient Legion
“following the guidance of military law” in II.4 ff. may also derive in some measure from
Patemus.

^Cf. in general for training R.W. Davies, “Fronto, Hadrian and the Roman army”. Service
in the Roman Army, ch. III, 71-90.

^militaris gradus.

^For the technical sense of the verb ambulo = decurro, cf. A.R. Neumann, “ROmische
Rekrutenausbildung im Lichte der Disziplin”, CPh xlii (1947) 157 n.3.

H.c. 18 miles, 658 yds. (29-57 km.) in 5 x ‘‘‘ Vij hr., or about 3-04 m.p.h. (4-89
km./h.). The Romans counted 12 hours of daylight, which averaged about H'A
modem hours in summer and 9'A in winter at the latitude of Rome. 1 standard
Roman mile measured c.1,617 yds. (1,478*5 m.). Cf. F. Hultsch, Griechische und
romische Metrologie (Berlin 1882^) 98, and W, Kubitschek, Grundriss der Antiken
Zeitrechnung, Handbuch der Altertumswissenschaft 1.7 (Munich 1928) 183.

^plenus gradus.

BOOK I: RECRUITMENT AND TRAINING

time.' If you add anything to this, it now becomes running, for which
a distance cannot be defined. But recruits should particularly be
accustomed to running too, so as to charge the enemy with greater
impetus, occupy favourable positions swiftly when need arises and seize
them first when the enemy wish to do the same, to go out on scouting
expeditions speedily, return more speedily, and overtake fugitives with
greater ease.^

The soldier should also be trained at jumping, whereby ditches are
vaulted and hurdles of a certain height surmounted, so that when
obstacles of this kind are encountered he can cross them without effort.
Furthermore, in the actual conflict and clash of arms the soldier coming
on by a running Jump makes the adversary’s eyes flinch, frightens his
mind and plants a blow before the other can properly prepare himself
for evasive or offensive action.^ Of the training undertaken by Gneus
Pompeius Magnus, Sallust records: “He would compete with the speedy
at jumping, with the swift at running, and with the strong at fencing.”''
For he could not otherwise have been a match for Sertorius, had he not
prepared himself and his soldiers for war with frequent exercises.^

10. Recruits should be trained in the art of swimming.

Every recruit without exception should in the summer months learn the
art of swimming, for rivers are not always crossed by bridges,^ and
armies both when advancing and retreating are frequently forced to
swim. Torrents often tend to flood after sudden falls of rain or snow,
and ignorance of swimming incurs risk not only from the enemy but the
water also. Therefore the ancient Romans, who were trained in the

'i.e. 22 miles, 86 yds. (35-48 km.) in 5 x'^ V ,2 hr., or about 3.65 m.p.h. (5-87 km/h,).
Cf. Livy 9.45.15 pleno gradu, cf. 30.5.4 modico gradu. Sail. Jug. 98.4 pleno gradu.
The militaris gradus is unique to V, It will correspond to the speed of the “standard
march” or iustum iter known from Caesar Bell. Ci\. 3.76, defined as a normal route-march
on good roads in good weather between camps, leaving time to build the camp and curare
corpora, and leave in good time the next day; cf G. Veith, in J. Kromayer, G. Veith
(edd.), Heerwesen und Kriegfiihrung der Chechen und Romer (Munich 1928) 352.

^Cf. Onas. 10.1-6.

’Cf 11.23, and Serv. ad Verg. Aen. 11.284, quoted ad loc.

‘‘Sail. Hist. frag. 2.19 (Maurenbrecher).

^Q. Sertorius is something of a hero with V., which supports a Spanish origin for him.
Cn. Pompeius Magnus held command as proconsul of Hither Spain against Q. Sertorius
from 77-71 BC.

sardonic allusion to urbani milites, modem field army men who were used to being
garrisoned in imperial capitals and, spoiled by the amenities of civilization, would
allegedly be unready to wade through rivers; cf III.8 init.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

whole art of warfare through so many wars and continual crises,
selected the Campus Martius next to the Tiber in which the youth might
wash off sweat and dust after training in arms, and lose their fatigue
from running in the exercise of swimming,’ It is highly advantageous
to train not just infantry but cavalry and their horses and grooms, whom
they call galearii^ to swim as well, lest anything should happen to them
on account of their inability in the hour of need.^

11. How the ancients trained recruits with wicker shields and with
posts.^

The ancients, as one finds in books, trained recruits in this manner.
They wove shields from withies, of hurdle-like construction, and
circular, such that the hurdle had twice the weight that a government
shield^ normally has. They also gave recruits wooden foils likewise of
double weight, instead of swords. So equipped, they were trained not
only in the morning but even after noon^ against posts. Indeed, the use
of posts is of very great benefit to gladiators as well as soldiers. Neither
the arena nor the battle-field’ ever proved a man invincible in armed
combat, unless he was judged to have been thoroughly trained at the
post. Each recruit would plant a single post in the ground so that it

'Cf. 1.3.

^See under 111,6. Cf. Jos. Bell !ud. 3.69 for the military training of soldiers’ servants.

’Arguably an allusion to the calamitous crossing of the Tigris by Jovian’s army in AD
363, cf. Amm. 25.8.1: aut imperitia nandi gurgite Jluminis absorbebantur... “or were
sucked under by the river’s current through tfieir inability to swim.” V. uses hyperbaton
to add pregnancy to his words quid imperitis... e\eniat\ cf. his allusion to the battle of
Adrianople, III. 11 fin.

'‘Cf Scipio Africanus Maior at Carthago Nova, 209 bc, Polyb. 10.20, Livy 26.51.3 ff. V.
omits the next stage, fencing with a live opponent, cf Antyllus apud Oribasius de Rem.
6.36.2—4. Cf Juv. 6,247 for use of posts in training gladiators. Cato de Re Mil. fr. 14 also
mentions gladiatorial training, so that P. Rutilius Rufus, cos. 105 BC, will not have been
the first to introduce gladiatorial methods of training to the army, despite Val. Max. 2.3.2.
Cf also Auct. de Bell Afr. 71 and Amm. 16.12.49.

^scutumpublicum. V. inserts a “modernizing” gloss having reference, probably, to the late
Roman army, in which arms issued to soldiers by the armorum custos remained state
property, subject to be handed back, cf refundere., 1.20 init. See M.P. Speidel, “The
Weapons Keeper, the fisci curator, and the ownership of weapons in the Roman army”,
Roman Army Studies II (1992) 131-136.

^V.’s consistent emphasis on afternoon training as something remarkable suggests that
Cato’s army did not observe the siesta, already a Roman custom by the first century BC,
when it was called the meridiatio.

^campus. The meaning varies according to context: 1.11, 11.25, III.15, 17, 20, 24, 25,
IV.45 (battle-field); 1.18, 26, II.7, 22, 23, III.2 (training-field); 1.27, III, 18, III.2, 6, 9,
19, IV. 1 (plain); 1.22, III.8 (camping ground).

BOOK I: RECRUITMENT AND TRAINING

could not move and protruded six feet. Against the post as if against an
adversary the recruit trained himself using the foil and hurdle like a
sword and shield, so that now he aimed at as it were the head and face,
now threatened the flanks, then tried to cut the hamstrings and legs,
backed off, came on, sprang, and aimed at the post with every method
of attack and art of combat, as though it were an actual opponent. In
this training care was taken that the recruit drew himself up to inflict
wounds without exposing any part of himself to a blow.

12. Recruits should be taught to strike not with the edge, but with
the point.

Further, they learned to strike not with the edge, but with the point. For
the Romans not only easily beat those fighting with the edge, but even
made mock of them, as a cut, whatever its force, seldom kills, because
the vitals are protected by both armour and bones. But a stab driven
two inches in is fatal; for necessarily whatever goes in penetrates the
vitals. Secondly while a cut is being delivered the right arm and flank
are exposed; whereas a stab is inflicted with the body remaining
covered, and the enemy is wounded before he realizes it. That is why,
it is agreed, the Romans used chiefly this method for fighting.’ The
hurdle and foil of double weight they gave out so that when the recruit
took up real, and lighter arms, he fought with more confidence and
agility, as being liberated from the heavier weight.

13. Recruits should be taught armatura.

Further, the recruit should be taught the type of drill known as
armatura^ handed down by drillmasters.^ This practice even partly

’The real reason was the introduction of the Spanish gladius, a short stabbing sword,
during the Second Punic War (probably on occasion of the reforms of Scipio Africanus
at Ca^ago Nova in 209 BC, cf Suda M.302 s.v. paxaipa, following a lost portion of
Polyb. book 29); but V.’s medical explanation is surely part of the reason for the change,
which made close-order tactics possible. An upward stab wound is far more dangerous
that a slash. The terms caesa, puncta are morphologically late forms, cf. A. Souter, A
Glossary of Later Latin to 600 AD (Oxford 1949) s. vv.

^A “special drill” attested as such only in the late-Roman sources Ammianus, cf Firmicus
Matemus Math. 8.6.3, and V., see TLL s.v. 606. But cf Livy 44.9.2-7 for similar displays
of manoeuvres and tactics by Roman youths in the Circus already commonplace by 169
BC, as under the Empire, cf 11.23. The equestrian display of the lusus Troiae is a related
custom which lasted centuries. Armatura pedestris was practised by the Emperors Julian
and Constantius II, and Ursicinus’ sons’ expertise in it could be represented as a threat to
Constantius II, cf Amm. 16.5.10, 21.16.7, 14.11.3. V.’s Emperor is praised for his
armatura (equestris) at 111.26 fin., as probably too is Constantius II at Amm. 21.16.7.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

survives. For it is well-known that even now armaturae' fight better
than the rest in all battles. Hence it should be appreciated how much
better is a trained soldier than an untrained, when armaturae, whatever
their proficiency, outstrip the rest of their comrades in the art of
warfare.

The discipline of training was so strictly maintained by our ancestors
that weapons instructors^ were rewarded with double pay, and soldiers
inadequately proficient in this mock-battle were forced to accept barley
instead of com. Wheaten rations were not restored to them until they
had shown, by giving demonstrations in front of the Prefect of the
legion,^ tribunes and senior officers,'* that they had completed all that
was contained in the military art.^ For there is nothing stabler nor more
fortunate or admirable than a State which has copious supplies of
soldiers who are trained. For it is not fine raiment or stores of gold,
silver and gems that bend our enemies to respect or support us;^ they are
kept down solely by fear of our arms. Secondly, in other matters, as
Cato says, mistakes can be corrected afterwards; errors in war do not
admit of amendment, because the penalty follows immediately upon the
slip. For those who fight without courage or expertise either perish

Claud, pan. de Hon. cons. VI 621-640 describes it. Also called pyrrhicha militaris, the
drills were set to music. See Grosse, 221-222.

^campidoctores, first attested in the Praetorian Guard under Septimius Severus, were
senior NCOs in late-Roman infantry units. CIL 2.4083 from Tarraco may attest a
centurion acting as campidoctor of the provincial governor’s equites singulares already
under Commodus.

^Armaturae may stand here for all scholares though properly designating scholae units
with this name only. Scutarii is similarly and more often used as a generic term, cf.
Hoffmann, 292. Scholae palatinae were 6lite mounted regiments belonging to the imperial
guard, cf. Jones, 613-614, Grosse, 93-96. V. has made a false etymology in all
probability, as there is no known connexion between the name of the regiments and that
of the drill which is likely to be much older, as stated above.

^doctores armontm.

^praefectus legionis, cf. II.9.

^principia, cf. 11.7.

^V. modernizes the officers and may invent the occasion. In the early Empire, at least,
the drill was taught by centuriones evocati acting as exercUatores, cf Ruggiero s.v.
Barley-rations as punishment may not have survived after Augustus, cf Suet. Aug. 24,
Front. Strat. 4.1.25, 37, Livy 27.13.9, Polyb. 6.38.3. They are absent from the milit^
punishments mentioned by the jurists Ulpian, Modestinus, Paul, Aem. Macer and Arrius
Menander, cf Dig. 49.16. Double pay for the higher NCO grade of armaturae is
mentioned at II.7, where V. may misidentify these with doctores armorum.

^’i.e., subsidies to barbarians.

BOOK I: RECRUITMENT AND TRAINING

immediately, or turn to flight and dare not stand up again to their
victors.’

14. Recruits should be trained at throwing javelins.

But I return to my subject. The recruit who is being trained with the foil
at the post is also made to launch spear-shafts of heavier weight than the
real javelins will be against the above-mentioned post as though against
a man. In this activity the instructor in arms sees to it that he hurl the
spear-shaft with great force, and drive the missile with a true aim into
or next to the post.^ For by this exercise the arms gain strength, and
skill and experience in throwing javelins is acquired.

15. Recruits should be thoroughly trained in arrows.

About a third or a quarter of recruits, who prove to have more aptitude,
should be trained constantly with the aforesaid posts using wooden bows
and mock arrows. Instructors should be chosen for this training who are
experts, and greater care should be taken that (recruits) hold the bow
scientifically, string it smartly, keep the left hand firm, draw the right
with calculation, let the eye and mind concentrate together on the target
to be hit, and learn to shoot straight whether from horseback or on foot.
This art needs to be learned thoroughly and maintained by daily use and
exercise. How much utility good archers have in battle was clearly
shown by Cato in his books On Military Science, and by Claudius, who
overcame an enemy to whom he had previously been unequal by
establishing and training numerous darters.^ Scipio Africanus, about to
do battle with the Numantines who had sent armies of the Roman

'Cf. III. 10 fm.

^Probably V.’s guesswork. One would not be permitted to miss the mark, surely, and in
any case the doctor armorum sounds suspiciously like the late-Roman campidoctor. But
an early doctor cohortis is attested among the Praetorian Guard probably in the second
century ad, who was promoted campidoctor of cohors I praetoria in CIL 6.533. Doctor
armorum, at least, does not seem to occur in inscriptions, cf. Ruggiero, s.v.

^M. Porcius Cato the Censor’s de Re Militari, written in the mid-second century BC, is
lost apart from a few fragments, collected by H. Jordan, M Catonis praeter librum de re
rustica Quae Extant (Leipzig i867). There is no reason to think V. had access to the
original. Appius Claudius Pulcher is probably credited here with the invention of velites
or light-armed sh^-shooters at the siege of Capua in 211 BC, but his proconsular
colleague Q. Fulvius Flaccus is usually more prominent in the tradition, cf Val. Max.
2.3.3, Livy 26.4.4-10. V. seems to be thinking of both archers— sagittarii —and
darters— iaculatores.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

People under the yoke, supposed that he would only get the upper hand
if he incorporated picked archers in all centuries.'

16. Recruits should be trained at firing stones from slings.^

It is advisable that recruits be thoroughly trained at casting stones by
hand or with slings. The inhabitants of the Balearic Isles are said to
have been first to discover the use of slings and to have practised with
such expertise that mothers did not allow their small sons to touch any
food unless they had hit it with a stone shot from a sling.^ Often,
against soldiers armed with helmets, cataphracts"* and cuirasses, smooth
stones shot with a sling or “sling-staff’^ are more dangerous than any
arrows, since while leaving the limbs intact they inflict a wound that is
still lethal, and the enemy dies from the blow of the stone without loss
of blood.^ That slingers served in all battles of the ancients is known to
everyone. This weapon should be learned by all recruits with frequent
exercise, because it is no effort to carry a sling.^ It often happens too
that warfare is carried on in stony places, that some mountain or hill has

'The sack of Numantia, Spain, 133 bc, by P. Cornelius Scipio Africanus Aemilianus, after
several years of war. Cf. Front. Strat A.121: Scipio Aemilianus ad Numantiam omnibus
non cohortibus tantum, sed centuriis sagittarios et Junditores interposuit. “Scipio
Aemilianus at Numantia interspersed archers and slingers not just in all cohorts, but in all
centuries.”

^Cf W.G. Griffiths, “The Sling and its place in the Roman Imperial Army” in 5th Roman
Military Equipment Conference, ed. C. van Driel-Murray, BAR Int. Ser. 476 (1989)
255-280, T. VOlling, “Funditores im rOmischen Heer”, SJ xlv (1990) 24-58.

^Cf. Flor. 1.43.5: (Baleares) certos esse quis miretur ictus, cum haec sola genti arma sint,
id unum ab infantia studium? Cibum puer a matre non accipit, nisi quern ipsa monstrante
percusserit. “Who can wonder that (the Baleares) are sure of aim, when this is their only
national arm, this their sole interest from infancy? A boy does not receive food from his
mother, unless he shall hit it when she points to it.” V. provides a colourful exemplum
here, perhaps again because of his national pride as a Spaniard.

^cataphractae, coats-of-mail or scale-armour, metallic or horn on a leather foundation, cf.
IV.9 fin. V. uses the term interchangeably with lorica —“cuirass”—, cf. 1.20, or else in
addition to it, as here.

^fustibalus. A kind of sling attached to a staff, also known in the Middle Ages, but
introduced in the late-Roman period. Cf T.G. Kolias, Byzantinische Waffen (Vienna 1988)
255.

^Probably a reminiscence of David and Goliath, cf Bibliavulgata/. Sam. 17.23,49, since
the idea is militarily unsound. V. pads out and embellishes from his own medico-Christian
perspective. Note the medical explanation.

W. is sarcastic about the modern, “urban” soldier’s ability to carry weights, cf 1.19.1.20.

BOOK I: RECRUITMENT AND TRAINING

to be defended, or barbarians kept from storming forts or cities,' by
using stones and slings.

17. On training ^ith lead-weighted darts.^

Training with lead-weighted darts, which they call mattiobarbuli, should
also be provided to recruits. In Illyricum once there were two legions
which had 6,000 men apiece and were called Mattiobarbuli after their
skilful and brave handling of these weapons. By them, as is well-known,
long ago all wars were concluded in a most vigorous manner; so much
so, in fact, that when Diocletian and Maximian acceded to the throne
they decreed that these Mattiobarbuli be called Joviani and Herculiani
in recognition of their valour, and are (thereby) judged to have preferred
them to all other legions.^ They usually carried five mattiobarbuli each,
slotted inside their shields. If soldiers throw them at the right moment,
it seems almost as if shield-bearing infantry" are imitating the role of
archers.^ For they wound the enemy and his horses before they can get
not merely to close quarters, but even within range of javelins.

^Civitas or “city” frequently means a small fort or fortified settlement in late-Latin, cf
Goffart, 65-66, but it is probably neutral here.

^(Sagittae) plumbatae, known to the elder Pliny Hist. Nat. 10.97, 12.85 after Arist. Hist.
An. 9,616 simply as “arrows weighted with lead”, as used in hunting for nests of the
cinnamon bird when they were shot from a bow, may have originated in military use as
the oestrus, a short wooden dart with an iron head, thrown from a sling, described by Livy
42.65.9 and Polybius 27.11 (9).l as having been first introduced c. 171 BC. From the
Anon. De Rebus Bellicis 10 we find it developing into a kind of throwing-mace in late
antiquity, also attested as such in Byzantine times, its weight increasing die while. But
they came in a wide variety of shapes and sizes from the late-Roman period, some clearly
darts for firing from engines, others for throwing by hand. There are no earlier examples
so far discovered by archaeology. Cf T. VOlling, “Plumbata-mattiobarbulus-
papT^^ofidt ppouXov. Bemerkungen zu einem Waffenfund aus Olympia”, AA (1991)
287-298, J. Bennett, “Plumbatae from Pitsunda (Pityus), Georgia, and some observations
on their probable use”, JRMES ii (1991) 59-63, R. Degen, “Plumbatae: Wurfgeschosse
der Spatantike”, HA xxiii (1992) 139-147.

^V. is the only source to attest the use of lead-weighted darts by the Joviani and
Herculiani, the most senior pair of legiones palatinae of the fourth century AD, cf
Hoffmann, 1.215-217. See also T. Drew-Bear, “Les voyages d’Aurdlius Gains, soldat de
Diocl^tien”, in La Geographie administrative etpolitique d’Alexandre a Mahomet, Actes
de Colloque de Strasbourg 14-16 juin 1979 (Univ. de Strasbourg 1982) 97, 101-102.
Viet. Caess. 39.18 implies that they were originally auxilia, a statement which may arise
from confusion with the Jovii. Diocletian acceded to the throne in ad 284 and Maximian
as his colleague in 286, and they created a number of units named after their divine
surnames “Jovius” and “Herculius”. For late-Latin doceantur, “are judged”, cf III.6 fm.
note.

^scutati, cf. 11.15.

^sagittarii.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

18. How recruits should be trained in mounting horses.

The vaulting of horses has always been rigorously demanded of both
recruits and of soldiers with many years’ service. It is clear that this
exercise has survived to the present day, albeit now with neglect.'
Wooden horses were set up under cover in winter, in summer on the
training-field.^ Over them recruits were made to vault at first unarmed,
until they had gained experience, then in arms. Such care was taken that
they learned to jump on and off both from the right-hand side and from
the left, even holding drawn swords and lances.^ This they likewise
performed with constant practice naturally so as to be able to mount
without delay in the stress of battle, having trained with such dedication
in peace.

19. Recruits should be trained in carrying a burden.

Recruits should very frequently be made to carry a burden of up to
60 lb."* and route-march at the military step, since on arduous campaigns
they have necessarily to carry their rations together with their arms.^
This should not be thought hard, once the habit has been gained, for
there is nothing that continual practice does not render very easy. We
know that the ancient soldiers used to do this exercise from the evidence
of Vergil himself, who says:

'Perhaps as part of armatura equestris, cf. Amm. 16.5.10, 21.3.7, Claud, pan. de Hon.
cons. VI 621-639, Grosse, 222. Polyb. 6.25.4 suggests that the vaulting of wooden and
live horses while holding arms was originally devised for velites in the third century BC.

^Cf. 11.23, III.2, Cf R.W. Davies, “Roman Military Training Grounds”, in E. Birley, B.
Dobson, and M.G, Jarrett (edd.), Roman Frontier Studies 1969, Eighth international
congress of Limesforschung (Cardiff 1974) 20-26, id., “The training grounds of the
Roman cavalry”, Service in the Roman Army, ch. IV, 93-124.

'Cf Arr. Tact. 43.3^.

C.43 lb. 5 oz. avoirdupois, 19-647 kg. 1 Roman pound = 0-7219 lb. avp. or 327-45
grammes, cf F. Hultsch, Griechische und romische Metrologie (Berlin 1882^) 161.

^Cf Amm. 17.9.2: ...ex annona decern dierum et septem, quam in expeditionem pergens
vehebat cervicibus miles, portionem subtractam^ “...a portion removed from the rations for
17 days, that a soldier going on expedition carried at his neck...” SHA Alex. Sev. 47.1:
...nec portarent cibaria decern et septem, ut solent, dierum nisi in barbarico, “...nor would
they carry rations for 17 days, as is normal practice, except in barbarian territory.” Front.
Strat. 4.1.7: C. Marius recidendorum inpedimentorum gratia, quibus maxime exercitus
agmen oneratur, vasa et cibaria militis in fasciculos aptata fitrcis inposuit, sub quibus et
habile onus etfacilis requies esset: unde etproverbium tractum est "mult Mariani”. “To
cut down on baggage, which weighs an army column down more than anything else, C.
Marius had the individual soldier’s cups and rations tied up in bundles and put upon
forked props, under which the burden might be manageable and respite easy: hence the
phrase passed into proverb, ‘Marius’ mules’.” See N. Fuentes, “The mule of a soldier”,
JRMES ii (1991) 65-100.

BOOK I: RECRUITMENT AND TRAINING

“Just as the bold Roman in his national arms

Cruelly laden takes the road, and before

The enemy expects it stands in formation, having pitched camp.’”

20. What kind of arms and armour the ancients used.^

The place (in our narrative) demands that we attempt to state the kind
of arms with which recruits should be equipped and protected. But on
this subject ancient practice has been utterly destroyed. For despite
progress in cavalry arms thanks to the example of the Goths, and the
Alans and Huns,^ the infantry as is well-known go unprotected. From
the founding of the City down to the time of the deified Gratian,"^ the
infantry army was equipped with both cataphracts^ and helmets. But
upon the intervention of neglect and idleness field exercises ceased, and
arms which soldiers rarely donned began to be thought heavy. So they
petitioned the Emperor that they should hand in^ first the cataphracts,
then helmets. Thus with their chests and heads unprotected our soldiers
have often been destroyed in engagements against the Goths through the
multitude of their archers. Even after so many defeats, which led to the

'Verg. Georg. 3.346-348.

^V.’s rhetorical allegation that infantry (legionaries) had cast off heavy armour has for
long been taken at face value and generalized incautiously. In fact it plays on a
commonplace used also by Tac. Ann. 13.35.1 and Pronto Princ. Hist. 11-12. For a
discussion of the evidence for generally undiminished use, at least in the East, of heavy
armour, cf. J.C.N. Coulston, “Later Roman armour, 3rd.-6th. centuries AD”, JRMES i
(1990)139-160.

^Cf. E.A. Thompson, A History ofAttila and the Huns (Oxford 1948) 53 on the Hun bow,
and V. Mul III.6 on the Hunnic horse.

'^Reigned in the West, AD 375-383, murdered at Lugdunum (Lyons) by Andragathius on
the order of Magnus Maximus, who invaded from Britain leading the regular troops
against Gratian, and exploited their hatred of him on account of his favouritism towards
his Alan mercenaries, cf. Zos. 4.35.2-6 = Eunapius fr. 51 (Blockley). Maximus reigned
as Augustus in the prefecture of Gaul until defeated and killed by Theodosius I at the
battle of Aquileia, AD 388.

^cataphractae, coats-of-mail or scale-armour, metallic or horn on a leather foundation, cf.
IV.9 fin. V. uses the term interchangeably with lorica —“cuirass”—, cf. below, or else in
addition to it, as at 1.16.

follow Otto’s emendation se de<be>re refundere. For refundere, cf SHA Claud. 14.5:
loricam unam, quant refundat. The circumstantial detail of a petition—cf No\. Theod. II
21 (441) for such a procedure followed by the scholae palatinae —may suggest an
historical event, but its scope is wholly a matter of conjecture. V. probably exaggerates
and generalizes a particular case, but the point was to stress the broken-backed state of
Gratian’s field infantry.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

sacking of so many cities,’ no one has troubled to restore either
cataphracts or helmets to the infantry.

The result is that those who in battle are exposed unprotected to
wounds, think not about fighting but fleeing. For what is a foot-archer
to do, without a cataphract or a helmet, when he cannot hold a shield
along with a bow? What are the dragon-bearers^ themselves and the
standard-bearers to do in battle, who control the poles with the left
hand, and whose heads and chests are obviously unprotected? But a
“cuirass”^ or helmet seems heavy to an infantryman who perhaps rarely
exercises, perhaps rarely handles arms; yet daily use makes light work
even if heavy equipment is worn. But these men, not being able to
endure the labour of wearing the old protective armour, because their
bodies are uncovered are forced to risk wounds and deaths, and worse
still, to be captured or betray the State by taking flight. So while they
refuse training and hard work, they are butchered in the greatest
disgrace, like cattle."*

Why else was the infantry army called a “wall” among the ancients,^
if not because the serried ranks of legions shone in their shields,
cataphracts and helmets? So much so, indeed, that archers were
equipped on the left arm with an arm-guard, and the shield-bearing
infantry were made to wear iron greaves on the right leg, as well as
cataphracts and helmets. Such was the armour of those who, fighting in

'I translate tantarum as = tot, as often in late Latin, and see no reference to the sack of
Rome. V. talks rhetorically of imumerabiles urbes at IV praef. The cities are likely to
have been in the Danube-tands, including such famous fortresses as Camuntum,
abandoned between ad 375 and c.390, cf. Amm. 30.5.2.

^In the Roman army the dragon was originally an auxiliary cavalry standard, described
as Scythian in a detailed account by Arr. Tact. 35.2-4, but it first appears on Trajan’s
column in use among the Dacians, cf. Cichorius PI. XXIII, scene xxxi, PI. XLV, scene
Ixiv, PI. LIX, scene cxliv, PI. LXIV, scene clix. Later it spread to the whole Roman army,
including legionary infantry units, by the fourth century ad, cf Grosse, 231-232, and see
J. Garbsch, Romische Paraderustungen (Munich 1978) 88 and PI. 48 for an actual dragon-
standard found at Niederbieber on the upper German limes. See also M.P. Speidel, “The
Master of the Dragon Standards and the Golden Tore”, TAPhA cxv (1985) 283-287, and
J.C.N. Coulston, “The ‘draco’ standard”, JRMES ii (1991) 101-114.

^lorica, = cataphract, cf above.

‘‘Cf Sail. Jug. 44.1, 44.5, Tac. Ann. 13.35.1, Pronto Principia Historiae 11-12, Pronto
at/Pen/m/mp. 2.1, echoed by contemporary sources such as Amm. 16.2.1, 18.6.2, 18.8.2,
22.4.6-7, 27.9.1, 27.9.6, 29.4.6, Liban. Or. 2.38, 47.6, SHA Tac. 2.4, Symm. Epp. 6.73,
7,38, Claud. In Eutrop. 2.580 ff Cf R. MacMullen, Corruption and the Decline of Rome
(Yale 1988) 175-176. Cic. Tusc. Disp. 2.16.37 gives the ideal.

^V. may have in mind a Latin translation of Horn. II. 4.299 epKoe; 7toX.ejioio, “bulwark
of war.” Cf 11.17; also Amm. 16.12.49: instar turrium, “like towers.”

BOOK I: RECRUITMENT AND TRAINING

the First line, were called principes,^ in the Second line hastati,^ and in
the Third line triarii.^ The triarii would wait in reserve behind their
shields on bended knees, lest they be wounded while standing by
oncoming missiles and, when circumstances required, would attack the
enemy with more violence for having rested. They often gained the
victory, it is said, after the hastati and those who stood at the front had
fallen."*

However, there were also in ancient times men among the infantry
called “light armament”, slingers, and light-armed troops.^ They were
mainly located on the wings, and undertook the opening phase of the
fighting. They were chosen for their great speed and fitness; they were
not very numerous. If the course of battle compelled it, they would

'“frontlinesmen”.

^“spearmen”.

^“thirdlinesmen”. These categories are from the manipular army of the third and second
centuries BC, essentially that described at greater length by Polyb. 6.22-23, and Livy 8.8
(which projects a manipular army back into the age of the Latin wars), and by V. at
11.15-17, 111.14. V. greatly exaggerates the use of cataphract-armour, under late-antique
influence, for it was worn originally only by the wealthiest soldiers, cf Polyb, 23.14-16.
He also inverts the canonical order hastati-principes of the First and Second lines,
probably on etymological grounds; in practice, too, they could really be inverted, cf Livy
8.8.9-10, 22.5.7, 40.27.6.

'’Cf. Livy 8.8.9-13 (allegedly 340 BC): Ubi his ordinibus exercitus insiructus esset, hastati
omnium primi pugnam inibant. Si hastati projligare hostem non possent, pede presso eos
retro cedentes in intervalla ordinum principes recipiebant. Turn principum pugna erat;
hastati segue bantur; triarii sub vexillis considebant, sinistro crure porrecto, scuta innixa
burner is, hastas sube recta cuspide in terra fixas, baud secus quam vallo saepta inhorret
acies, tenentes. Si apud principes quogue baud satis prospere esset pugnatum, a prima
acie ad triarios se sensim referebant. Inde rem ad triarios redisse, cum laboratur,
proverbio increbuit. Triarii consurgentes, ubi in intervalla ordinum suorum principes et
hastatos recepissent, extemplo compressis ordinibus velut claudebant vias unoque
continenti agmine, iam nulla spe post relicta, in hostem incidebant; id erat
formidolosissimum hosti, cum velut victos insecuti novam repente aciem exsurgentem.
auctam numero, cernebant. “When the army had been drawn up in these lines, the hastati
began the battle first of all the rest. If the hastati could not overcome the enemy, falling
back at the double they were received between intervals in the lines of the principes. Then
the battle belonged to the principes; the hastati were behind them; the triarii sat under
their insignia, with the left leg forward, holding their shields rested against the shoulders,
their spears fixed in the ground with points lowered, Just as if the line were bristling with
a fence of stakes. If the principes also had insufficient success in the fight, they slowly
withdrew from the front line to the triarii. Hence it became proverbial to say that ‘the
thing was down to the triarii\ when in difficulty. The triarii rose up and received the
principes and hastati between Intervals in their ranks, and then suddenly closing ranks
they as it were shut off the pathways and made a single continuous line, whereupon with
no other hope left to fall back on they attacked the enemy. It was a most terrible thing for
the enemy when, pursuing apparently defeated men, they suddenly saw a new array rising
up against them in increased numbers.”

^ferentarii.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

retire and be received between the front ranks of the legions, in such a
way that the line remained unmoved. Almost down to the present day
the custom survived whereby all soldiers wore leathern caps which they
called “Pannonians”. This was observed so that the helmet should not
seem heavy to a man in battle, who was used always to wearing
something on his head.'

The javelins that the infantry army used were called pila, having thin,
triangular iron heads 9 in. or 1 ft. long, such as once lodged in a shield
could not be broken off and, when thrown skilfully and with force,
might easily pierce a cuirass.^ Weapons of this type are now rare with
us, but barbarian shield-bearing infantry use these particularly, calling
them bebrae, and carrying two or three each in battle.^ It should be
noted, also, that when missiles are being exchanged, soldiers should
have the left foot forward;"* this way the throw is stronger for hurling
darts.^ But when it comes to what they call “to Javelins”,^ and the
fighting is hand-to-hand with swords,^ soldiers should have the right
foot forward,* so as to draw the flank away from the enemy lest they be
wounded, and to have the right hand closer so it can land a blow.

Obviously, therefore, recruits should be equipped and protected with
every ancient type of arms. For a man who does not fear wounds
because he has his head and chest protected must acquire sharper
courage for battle.

‘V.’s medical exegesis again. The caps are thought to be those depicted in artistic
representations of soldiers in non-battle dress, familiar from the famous relief of the
Tetrarchs on St. Mark’s, Venice, cf. H. UbI, “Pilleus pannonicus, die Feldmiltze des
spatr5mischen Heeres”, in Festschrift fur R. Pittioni II (Vienna 1976) 214-241. Leathern
caps arc mentioned because the source mentioned the headgear of velites here, cf. Polyb.
6.22.3, RE VII (1910) 572 s.v. Galea (Fiebiger), J. Kromayer, G. Veith, Heerwesen (\92^)
327, D. Schenk, 36.

^Cf. under II.15.

'J.G. Kempf, “Romanorum sermon is castrensis reliquiae collectae et illustratae”,
Jahrbucher fur klassische Philologie, Suppl. xxvi (1901) 367, identifies this with a
hypothetical Celtic *berva = spear.

*in ante, a I ate-Latin ism, cf. E. LOfstedt, Late Latin (1959) 164 n.l.

^spicula^ cf. 11.15.

"'adpila.

^V. means the close-quarters phase of battle, caWed ad spathas et ad pila, ‘‘to broadswords
and to javelins” at III. 14.

^proximior, a late-Latinism quoted by Priscian Inst. 3.21.

BOOK I: RECRUITMENT AND TRAINING

21. On building camps.^

The recruit should also learn how to build camps, for nothing is found
so safe or so indispensable in war, since if a camp has been properly
constructed, soldiers spend days and nights so secure behind the
rampart—even if the enemy is besieging it—that they seem to carry a
walled city about with them everywhere.^ But knowledge of this
technique has altogether perished; for no one now for a long time has
built a camp with lines of fosses and a stockade fixed above it.^ For this
reason, as we know, many armies have frequently been harmed in day
or night-attacks by barbarian cavalry. But it is not only when they
bivouac without building a camp that they suffer this: when in battle
they begin to retreat for some reason, they lack the fortification of the
camp to retire to and fall unavenged, like animals, and there is no end
to the killing until the enemy’s will to pursue flags.

22. In what kind of places camps should be built.

Camps—especially when the enemy is near—should be built always in
a safe place, where there are sufficient supplies of firewood, fodder and
water, and if a long stay is in prospect choose a salubrious site. Care
must be taken lest there be a nearby mountain or high ground which
could be dangerous if captured by the enemy. Thought must be given
that the site is not liable to flooding from torrents and the army to

‘1.21-25 on castrametation are not well-adapted to the training of recruits but appear to
have been put together by V. and inserted incongruously into the training chapters. For
the recruit would actually need to acquire trenching, building and stockading skills,
mentioned at 1.4, 11.25, III.4, III. 10, whereas the strategic calculations which V.
summarizes here were relevant only to senior officers. The material is repeated with a few
variations and additions at III.8. Cf. R.W. Davies, “Roman Wales and Roman military
practice-camps”, Service in the Roman Army, ch.V, 125-139, and R. Grosse, “Das
rOmisch-byzantinische Marschlager vom 4.-10. Jahrhundert”, Byz.Z. xxii (1913)90-121,
arguing that the fourth-century army still built camps until growing barbarization lost the
technique.

^Cf 11.18, 11.25 fin. The mobile city is a peculiarly late-Roman concept, inspired by the
Gothic carrago which is cited as an example for Roman castrametation at 111.10. Note that
V. includes circles among possible shapes of camps at 1.23, III.8. But the image of the
camp as a city goes back at least as far as Polyb. 6.31.10, cf Front. Strat. 4.1.14.

rhetorical exaggeration, also at III. 10, but even dlite barbarian units such as auxilia
palatina do not appear usually to have had the same skills in castrametation or tolerance
of hard work (cf II.3) as the trained legions. Cf Amm. 18.6.2; auxiliarii milites semper
mania spernentes huiusmodi, “auxiliary soldiers who always refuse tasks of this sort”, and
19.5.2, but contrast 31.8.9.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

suffer harm in this event.' The camp should be built according to the
number of soldiers and baggage-train, lest too great a multitude be
crammed in a small area, or a small force in too large a space be
compelled to be spread out more than is appropriate.

23. In what shape camps should be marked out.

Camps should be made sometimes square, sometimes triangular,
sometimes semicircular, according as the nature and demands of the site
require. But the gate which is called praetoria^ should either face east,
or the direction which looks towards the enemy, or if on the march it
should face the direction in which the army is to proceed.^ Within it the
First centuries, that is cohorts, pitch tents and set up the dragons and
standards."* The gate called decumana^ is behind the headquarters;^
through it delinquent soldiers are taken out to be punished.

24. The sorts of fortification of a camp.

There are three different sorts of fortification of a camp. When there is
no pressing danger, turves are cut from the earth and from them a kind
of wall is built, 3 ft. high above the ground, with the fosse from which
the turves were lifted in front. Then there is a temporary fosse 9 ft.
wide and 7 ft. deep. But when more serious forces of the enemy
threaten it is advisable to fortify the perimeter of the camp with a
proper fosse, 12 ft. wide and 9 ft. deep below the “line”, as it is called.^
Above it revetments are built on either side* and filled with earth that
has been raised from the fosse, rising to a height of 4 ft. The result is
that (the fosse) is 13 ft. deep and 12 ft. wide. Above are fixed stakes of
very strong wood, which the soldiers are accustomed to carry with them.

'Cf, Ps.-Hyg. 57 for the same regulations. Caesar at Ilerda in 49 BC made the mistake of
camping between two rivers, Caes. Bell. Civ. 1.48.

^Main gate,

W. adds the probably Christian eastward orientation to the traditional regulations given
also by Ps.-Hyg. 56.

W. normally identifies centuries with cohorts, giving them dragon-standards, obviously
under the influence of late-Roman usage.

^Rear gate.
praetor in m.

^i.e. ground-level.

'*i.e. on either side of the intended embankment, which was on the inner side of the fosse,
cf. 111.8.

BOOK I: RECRUITMENT AND TRAINING

For this work it is advisable always to have in readiness mattocks, rakes,
baskets and other kinds of tools.'

25. How a camp should be built when under threat from the enemy.

It is easy to build a camp when the opposition are absent, but if the
enemy threatens, then all the cavalry and half the infantry are drawn up
in line to repel attack^ while the remainder build the camp behind them,
digging fosses. The herald announces which century is the first, which
the second and third, to have completed all its work. After this the fosse
is inspected and measured by the centurions, and punishment is inflicted
on any who may have worked carelessly.

Therefore the recruit should be trained in this technique, so that when
need arises he may be able to build a camp without difficulty, quickly
and in safety.

26. How recruits are trained to keep ranks and intervals in the line.

It is agreed that nothing is more advantageous in battle, than that by
dint of their constant exercises soldiers should keep their appointed
ranks in the line,^ and not mass together or thin out the formation at any
point inconveniently. For when densely packed they lose room to fight
and impede one another, and when too thinly spread and showing the
light between them they provide the enemy with an opening to breach.
It is inevitable that everything should at once collapse in panic if the
line is cut and the enemy reach the rear of the fighting men. Therefore
recruits should be led out constantly to the exercise-field and drawn up
in line following the order of the roll, in such a way that at first the line
should be single and extended, having no bends or curvatures, and there
should be an equal and regular space between soldier and soldier."^ Then
the command should be given that they at once double the line, so that
in an actual assault that arrangement to which they are used to conform
may be preserved. Thirdly the command should be given suddenly to

'Cf. 11.25. See N. Fuentes, “The mule of a soldier”, JRMES ii (1991) 65-100.

^Cf. 111.8, indicating that the cavalry and half the infantry were exempted by rank from
camp-construction.

^The traditional penalty for failure to do this was death, cf. Livy 2.59.11, Diod. Sic.
12.64.3. But while training his troops at Numantia in the 130s BC, Scipio Africanus Minor
punished such errors with flogging only, cf. Livy Epit. 57.

'Cf. 111.15.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

adopt a square formation,* and after this the line itself should be
changed to triangular formation, which they call a “wedge”.- This
formation is usually of great advantage in battle. Next they are
commanded to form circles,^ which is the formation commonly adopted
by trained soldiers to resist a hostile force that has breached the line, to
prevent the whole army being turned to flight and grave peril ensuing/
If recruits learn these manoeuvres by continual practice they will
observe them more easily in actual battle.

27. The distance that they should advance and retire, and how often
a month the exercise should take place, whereby soldiers are led out
on marching manoeuvres.

Next, the ancient procedure both survived and is enacted by
constitutions of the deified Augustus and Hadrian, that thrice a month
both cavalry and infantry should be led out ambulatum,^ That is the
term by which they call this type of exercise: the infantry were
commanded to advance ten miles at the military step, armed and
equipped with all their weapons, and then to retire to camp, completing
some part of the march at the brisker running pace.^ The cavalry also,
divided by troops’ and armed, similarly traversed the same distance,
performing equestrian manoeuvres, now pursuing, now retreating, and
by some rally renewing the charge.® Both formations had to go not
merely on the flat but to go down and up in hilly and difficult terrain,
to the end that nothing, even quite by chance, could befall the fighting

^quadrata acies, a marching formation with a hollow centre in which the baggage-train
could be escorted; cf. III.6.

^cuneus, in fact a marching formation formed from two converging lines, only
metaphorically like a triangle or wedge; cf. F. Lammert, “Der Keil in der Taktik des
Altertums”, Gymnasium li (1940) 15-31. Cf.IlI.17-20.

^orbes.

'Cf. Sail. Jug. 97.5, Caes. Bell. Gall. 4.31.2, 5.33.3, Amm. 29.5.41, 28.5.6.

^“on manoeuvres”. Every seven days, according to SHA Avid. Cass. 6.3; every five days,
according to SHA Maxim. 6.2. The present tense praecavetur appears to refer to the
constitutions of Augustus and Hadrian as an extant document, but is probably merely
copied from the source derived from Paternus; this is the only instance of V. citing
constitutiones, despite 1.8. For the technical sense of the verb amhulo = decurro, cf A.R.
Neumann, “ROmische Rekrutenausbildung im Lichte der Disziplin”, CPh xlii (1947)
157 n.3.

^’Cf 1.9 for the different paces.

^turmae.

"^Cf R.W. Davies, ‘The Training grounds of the Roman cavalry”. Service in the Roman
Army, ch. IV, 93-123.

BOOK I: RECRUITMENT AND TRAINING

men, which good soldiers had not learned to deal with beforehand by
the constant practice.'

28. On encouraging military science and Roman valour.

This material, having regard for faithfulness and devotion to truth, I
have brought together in this little book, Invincible Emperor, abridged
from all the authors who have written down the technique of military
science, so that should anyone wish to be diligent in raising and training
recruits he may be able easily to strengthen an army in emulation of
ancient military virtue. For martial energy has not declined in mankind,
nor are the lands exhausted that produced the Spartans, the Athenians,
the Marsians, Samnites, Paeligni, the Romans themselves.^ Were not
once the Epirotes very powerful in arms?^ Did not the Macedonians and
Thessalians overcome the Persians and penetrate as far as India on
campaign?'' Clearly the Dacians, Moesians and Thracians have always
been warlike, for fables tell us that Mars himself was bom among
them.^ But it is tedious if I attempt to list the strength of all the
provinces, when they all belong under the sway of the Roman Empire.^
However, a sense of security bom of long peace^ has diverted
mankind partly to the enjoyment of private leisure, partly to civilian

'Cf. Jos. Bell. lud. 3.98 on leaving nothing to chance.

^The Marsians and Paeligni, Italian tribes of the central Apennines, were the core of the
unsuccessful last great challenge to Roman power in Italy mounted by these and other
peoples, including the Samnites, during the Social War, 91-87 BC. Earlier Rome had
fought three major wars with the Samnites, a powerful Italian tribal confederation of the
southern Apennines, between 343 and 290 BC for hegemony in the Italian peninsula.

^A mountainous state abutting the NW Greek coast, Epiros became a Hellenistic military
great power based on its capital Ambracia, pushed back the empire of Macedon, and
annexed most of north Greece for a time under its formidable king, Pyrrhus. He reigned
297-272 BC, and among ceaseless campaigns fought the Romans in Italy and the
Carthaginians in Sicily, 280-275.

■’Under Alexander III, the Great, king of Macedon, reigned 336-323 BC, conquered and
overthrew the Persian Empire 331, went on expedition to India 327-325.

^The story of Mars’ birth is told only by Ovid Fasti 5.251-260.

^The list of provinces and peoples picks up that at I.l which began with the peoples of
the west before moving to Greece. The manpower reserves of the whole Empire are
meant. Hieron. Epp. 60.16.2 (ad 397) may possibly ironically echo this passage and the
20+ years of neglect mentioned below.

’Cf. 1.5, 1.7, II.3, II. 18, III. 10. V. is under the influence of the literary topos in Juvenal
6.292: nunc patimur longae pacts mala. “Now we are suffering the evils of long peace.”
He really means a lengthy period of neglect of the army. It is not clear how long the
“long peace” is envisaged as being, although it was certainly running in Gratian’s reign
(cf 1.20).

VEGETIUS’ EPITOME OF MILITARY SCIENCE

careers,' Thus attention to military training obviously was at first
discharged rather neglectfully, then omitted, until finally consigned long
since to oblivion.^ Neither let anyone wonder that this happened in the
preceding age,^ because after the first Punic war twenty and more years
of peace so enervated those all-conquering Romans as a result of private
leisure and neglect of arms, that in the second Punic war they could not
stand up to Hannibal. So it was that after so many consuls, so many
generals, so many armies lost, they only finally achieved victory, when
they had been able to learn military science and training."* Therefore
recruits should constantly be levied and trained. For it costs less to train
one’s own men in arms than to hire foreign mercenaries.^

‘Cf 1.5, III.10.

^Cf. 1.8, 1.20, and 1.21. The closing strictures on the failure to train parallel those on the
failure to recruit, 1.7. Cf also 11.3, II.18 fin., and III. 10 fin.

^Superiore aetate, i.e. the recent past, cf 111.12: superiore vel nostra aetate^ referring to
AD 378; so perhaps translate “the preceding reign”.

"•a ref to Scipio Africanus’ training programme at Carthago Nova, 209 BC, cf Polyb.
10.20, Livy 26.51.3 ff. The First Punic War lasted 264 -241 bc, the Second 218-201 BC.

^Cf refs, to expenses of the army at 11,1, II.3, 11.18, 111.3. See also Introduction fin.,
and 1.2,11.1-3, Ill.I, III. 10, obliquely referring to over-reliance on barbarians, as cavalry,
auxilia palatina, and as irregulars.

Book II

Preface.

It is demonstrated by continual victories and triumphs that Your
Clemency abides by our ancestors’ precepts for the armed forces to the
full, and with very great expertise, if indeed the outcome of events is an
undoubted proof of one’s ability. But Your Tranquillity, Invincible
Emperor, with loftier counsel than an earthly mind could have
conceived, requires ancient examples from books, whilst surpassing
antiquity itself by fresh deeds. Therefore when I was commanded to
collect this material in a written epitome for Your Majesty not so much
to learn as to recollect, devotion to the truth often contended with
embarrassment. For what could be more audacious than to address some
statement about the practice and science of warfare to the Lord and
Emperor of the human race, the Conqueror of all barbarian races, unless
of course he had ordered to be done what he had done himself?' Then
again, not to obey the commands of so great an Emperor seemed full of
sacrilege and danger.

So, in a wonderful way, I was rendered bold in obeying, by fearing
to appear still more bold by refusing. The previous indulgence of Your
Eternity encouraged me to this act of temerity. For I offered earlier in
all humility a booklet on the levying and training of recruits,^ and yet
escaped without blame. Nor am I afraid under orders to tackle a work
which, when it was voluntary, came off with impunity.

(The armed services, ch. 1-3)

1. Into how many categories military science is divided.^

Military science, as the most excellent of Latin authors bears witness in
the opening of his poem, consists of “arms and men”."* These are
divided into three parts, cavalry, infantry and navy. The cavalry
squadrons^ were so called because they covered the battle-line on each

‘The titles were applied to several fourth- and fifth-century Emperors, cf. A. Chastagnol
in A. Donati (ed.). La Terza Eta dell’ Epigrafia, Atti del Convegno, Univ. di Bologna,
ottobre 1986 (Faenza 1988) 24-25.

^preserved as Book I.

^A traditional subject in the tactici, defining the terms in which the science is organized;
cf. Arr. Tact 2 ff.

^alluding to Verg. Aen. I.l: arma virumque cano... “Arms and a man I sing...”

^alae^ “wings” = squadrons.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

side like wings;' they are now called “vexillations”" after the flag,^ for
they use flags, that is flame-coloured ensigns/ There is also another
type of cavalry, called “legionary” because they were joined to the
legion; “greaved” cavalry were established on their model/ There are
likewise two sorts of navy, one of warships,^ the other of river patrol-
boats/ Plains are held by the cavalry, seas and rivers by the navy, hills,
cities, flat and broken country by the infantry. From this it is clear that
infantry are more vital to the State, as they can be useful everywhere;
also a greater number of soldiers is maintained at less cost and expense/
The army^ received its name from the actual fact and action of
exercise, so that it was never permitted to forget what it was called/^
The infantry itself was divided into two parts, namely into auxiliary
units" and legions. The anxilia were supplied by the Allies'^ and
“federated peoples”;'^ whereas Roman manpower mainly predominated

'Cf. Cincius apud A. Ge!l. 16.4.6: alae dictae exercitus equitum ordines, quod circum
legiones dextra sinistraque tamquam alae in avium corporibus locabantur “The lines of
cavalry of an army were called ‘wings’, because they were stationed about the legions on
the right and left-hand sides just as wings on the bodies of birds.”

^vexillationes.

^velum^ i.e. vexillum.

^Jlammulae, see Ill. 5. Cf. Amm. 20.6.3: etmatutinae lucis exordio, signo per Jlammeum
erecto vexillum, circumvaditur civiias... “and at first light, as the signal of the flame-
coloured ensign was hoisted, the city was surrounded.” Cf Leo Tact. 5.4, 6.2.

^ocreati equites. They may be identical to Procopius’ mounted archers wearing cuirasses
and greaves, cf Procop. Bell. 1.1.8-16. Cf also Arr. Tact. 4.8 who may allude to their
ancestor. V. seems to imply that ocreati equites were created after the abolition of
legionary cavalry, i.e., in the early fourth century AD. Cf Grosse, 17, 36, 49.

^'liburnae.

^lusoriae.

is often accused of reacting against the late-imperial rise of cavalry, but in fact
regular cavalry units only made up one-third of the number of units listed for instance in
the eastern Notitia Dignitatum, cf Hoffmann 1.193 ff., Liebeschuetz 33. When V. talks
of the need for infantry, he means Roman legionary infantry, as opposed to barbarian
auxilia or vexillationes. He may have regarded both the latter as little better than the
barbarian mercenaries who also cost more (cf 1.28 fin.).

'^exercitus.

"'’Cf Varro de ling. lat. 5.87: exercitus, quod exercitando fit melior. “‘Army’ (exercitus),
because it is made better by means of ‘exercise’.”

''auxilia.

'^socii (Latin, Italic and other allies).

'foederatae gentes, i.e. civitates foederatae.

BOOK II: THE ANCIENT LEGION

in the ranks of the legions.' The legion was named after “selecting”,^
because their appellation requires the good faith and conscientiousness
of persons approving soldiers. In the auxilia a smaller number, in the
legions a far greater number of soldiers was usually enrolled.^

2. The difference between legions and auxilia."^

The Macedonians, Greeks and Dardanians^ had phalanxes, such that in
one phalanx 8,000 soldiers were counted. The Gauls, Celtiberians, and
many other barbarian nations used hordes^ in battle, in which there were
6,000 soldiers. The Romans had legions, in each of which there
traditionally were 6,000 servicemen, sometimes more.^ I shall explain
what the difference is between legions and auxilia. When auxiliaries are
taken on campaign, coming as they do from diverse stations and diverse
units, they have nothing in common with one another either in training,
in acquaintance, or morale. Their methods differ, the use of arms is
different among them. Further, those who disagree before they fight
must achieve victory more slowly. For instance, although it is of great
importance on campaign that all soldiers wheel at the bidding of a
single command,* those who were not together before cannot carry out
their orders evenly. But these same units are of no little assistance if
they are hardened by the various customary exercises almost daily. For
auxilia were always joined to the (old) legions in battle as light

'In the middle Republic there were legions of socii and Latins, as well as those of Roman
citizens; cf. Livy 37.39.7.

^eligendum. Cf, Varro de ling. lat. 5.87: legio, quod leguntur milites in delectu, cf. ibid.
6.66. “‘Legion’, because soldiers are ‘selected’ in the ‘selection’-process.”

^Referring to the smaller size of auxiliary as compared to legionary units. The rubric to
ch. 2 has perhaps become displaced and should rather go before Quid autem inter...
expediam —“I shall explain...”.

* Auxilia for V.’s generation meant elite barbarian light regiments of auxilia palatina, cf.
Hoffmann, 1.72. V. blames them for part of the disparateness of the modern field army.
But V.’s source was referring to the old auxiliary units which assisted the old legions in
a more subordinate r61e. V. surreptitiously uses criticisms of the old to attack the new
auxilia.

^The Dardanians were a warlike people of the central Balkans, N. of Macedonia, in what
is now the plain of Kosovo. Subdued by the Macedonians in 335 bc and by the Romans
in the first century BC, they proved difficult to control in practice, but were a fruitful
source of army recruits to both empires.

^'catervae.

^Cf. Isid. Etym. 9.3.46; proprie autem Macedonum phalanx, Gallorum caterva, nostra
legio dicitur. “Properly speaking, ‘phalanx’ is said of the Macedonian, ‘horde’ {caterva)
of the Gallic, and ‘legion’ of our own unit.”

’'Cf Jos. Bell lud. 3.88, 105.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

armament, so that they were used more as an aid to fighting than the
main reserve force.' But when the legion is fully complemented with its
own cohorts, when it has the heavy armament, that is principes, hastati,
triarii^^ antesignani,^ and also the light armament, that is ferentarii^^
sagittarii,^ funditores,^ ballistarii^ when it has its own legionary cavalry
incorporated in it and on the same rolls,* when it acts with one mind
and equal commitment to fortify a camp, draw up a line, do battle,
complete in every part and needing no external additions,—it usually
defeats any numbers of the enemy.^ Proof of this is the great size of the
Roman State, which always fought with legions and conquered as many
enemies as either it wished or the nature of the world permitted,'®

3. What caused the legions to become depleted.

The name “legion” survives in the army even today, but its strength of
former times has been broken by neglect, since corruption usurped the
rewards of valour and soldiers were promoted through influence when
they used to be promoted for actual work." Secondly, when ordinary
soldiers had completed their years of service and were duly discharged
with testimonials, others were not substituted. Also, inevitably some are

^subsidium. The auxilia palatina appear to have had just such a role, cf. Amm. 16.12.45
(Batavi and Reges, paired auxilia palatina), 24.6.9 (perhaps the same pair), 25.6.2-3
(Jovii and Victores), 31.13.9: Batavos in subsidiis locates, “the Batavians placed in
reserves.” The proper subsidium to which V. refers belonged to the triarii, cf. Varro de
ling. lat. 5.89: Pilani triarii quoque dicti, quod in acie tertio ordine extremi subsidio
depone bantur; quod hi subside bant abeo subsidium dictum, a quo Plautus: " A gite nunc,
subsidite omnes quasi solent triarii”. “The pilani were also called triarii, because they
were placed in the third line in reserve, last in the battle-array. Because they ‘sat down’
{subsidebant), the word ‘reserve’ (subsidium) was derived from it; hence Plautus: ‘Come
now, sit down all as the triarii do’.”

'Cf. 1.20.

^Cf II.7, 11.16, where they are considered akin to standard-bearers, drillmasters, NCOs.
‘’“light troops”, cf. 1.20.

'“archers”, cf 1.15.

"“slingers”, cf 1.16.

'“catapultiers and crossbowmen”, cf 11.15 fin., 111.14.

’’Cf 1.20, 11.15-17, III.14. This legion with its own light-armed derives from Cato’s lost
de Re Militari, but V. is unreliable on details and habitually updates some of the
weaponry.

‘^V. implies that the late-Roman field army, which was a hotch-potch of different types
of forces, was weakened by its composition; he is also criticizing the separate existence
of (mainly barbarian) light infantry and cavalry units.

‘"Cf Jos. Bell. lud. 3.107, Flor. 1.34, 19.1, 5.

"Cf Sail. Cat. 52.22: omnia virtutis praemia ambitio possidet, “corruption takes all the
rewards due to bravery.” Cf Amm. 20.5.7, Zos. 4.27.1-3.

BOOK II: THE ANCIENT LEGION

weakened by disease and are discharged, others desert or die for various
reasons;' an army becomes depleted, however numerous, unless every
year, or rather virtually every month, a host of recruits succeeds to the
places of those who have left. There is also another reason why the
legions have become attenuated: the labour of serving in them is great,
the arms are heavier, the duties^ more frequent, the discipline more
severe. To avoid this many flock to the auxilia^ to take the oaths of
service, where the sweat is less and the rewards come sooner.'*

Cato the Elder, since he was unbeaten in war and as consul had often
led armies, thought he would be of further service to the State if he
wrote down the military science. For brave deeds belong to a single age;
what is written for the benefit of the State is eternal.^ Several others did
the same, particularly Frontinus, who was highly esteemed by the
deified Trajan for his efforts in this field.^ These men’s
recommendations, their precepts, I shall summarize as strictly and
faithfully as I am able.^ For although both a carefully and a neglectfully

'Cf. Aem. Macer de Re Mil. II apud Dig. 49.16.13 tj3. Missionum generates causae sunt
tres: honesta causaria ignominiosa. Honesta est, quae tempore militiae impleto datur:
causaria, cum quis vitio animi vel corporis minus idoneus militiae renuntiatur;
ignominiosa causa est, cum quis propter delictum Sacramento solvitur. “There are three
general reasons for discharge: honourable, causal and ignominious. The honourable is that
which is granted on the completion of the period of military service. The causal is when
one is declared unfit for military service because of some fault of mind or body. The
ignominious reason is when one is released from the oath of service because of some
crime.” By a law of Constantine, soldiers in the field armies and the riparian legions were
permitted honesta missio after a minimum of 20 years’ service, but not the full privileges
of veterans— emerita missio —for which 24 years was the minimum; cf. Jones, 635.
^munera, fatigues, services to officers.

M.e., auxilia palatina.

comment on a failure to maintain the paper strength of the late-Roman legions, small
as they were, rather than on reductions in paper strength from the 6,000-strong old legion.
The same problem of unattractive conditions leading to vacancies is tackled by the Anon.
de rebus bellicis 5. On the lighter service of the auxilia palatina, cf Amm. 18.2.6;
auxiliarii milites semper munia spernentes huiusmodi ad obsequendi sedulitatem luliani
blanditiis deflexi... “the auxiliary soldiers, who always refuse labours of this kind, were
persuaded to complete obedience by the blandishments of Julian...’' with HotTmann, 1.72.
Note V.’s unmarked transition to modem auxilia.

^Cf Sail. Cat. 3.1. V. goes beyond Sallust’s argument that writing is as good as doing.

^’Sex. Julius Frontinus was consul ter ordinarius in AD 100 with Trajan. But his
Stratagems were written under Domitian. Possibly V. wished to hint that his own relations
with his Emperor had parallels with those between Frontinus and Trajan. See 1.8 note
above on Theodosius-Trajan.

^Cf 1.8 for the source-notice to book 1. This source-notice to book 11 seems to do duty
for II-IV, which were all published to commission to complement book I which was not.
The other sources to book I are subsumed under alii conplures here, so that the sources
in general are the same for all books.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

ordered army cost the same expense, it is to the benefit of not only the
present but of future generations also if, thanks to Your Majesty’s
provision, August Emperor, both the very strongest disposition of arms
be restored and the neglect of your predecessors amended.

(The formation and structure of the Ancient Legion, ch. 4—14)

4. How many legions the ancient generals led to war.

In all the authorities it is found that individual consuls led against the
most numerous hostile forces no more than two legions each, with
auxilia of the Allies added.* So great was their training, such their
confidence, that two legions were deemed sufficient for any war.^
Accordingly I shall explain the organization of the ancient legion
following the guidance of military law.^ If this description seems rather
obscure or incomplete, it is proper to blame not myself but the difficulty
of the actual subject. Therefore the details should be read repeatedly
with an attentive mind so that with memory and intelligence they may
be understood. For the State must inevitably be invincible, whose
Emperor, having learned the military art, may make as many armies as
he wishes fit for war.

5. How the legion is formed.

So when recruits have been carefully selected who excel in mind and
body, and after daily training for four or more months,"* a legion is
formed by order and auspices of the invincible Emperor. The soldiers
are marked with tattoos in the skin which will last and swear an oath,
when they are enlisted on the rolls. That is why (the oaths) are called

^socii (Latin, Italic and other allies).

^Cf. Polyb. 3.107.10-15, 6.19-20, Livy 22.36.2-4. The Romans first fielded more than
four legions (eight) at Cannae 216 BC.

militare is defined by Isid. Etym. 5.7.1-2 as rules of conduct of war in the widest
sense, including both international relations, army discipline in the field, and the
differentiation of pay, rank and rewards. But here it may simply allude to material derived
from the jurist P. Taruttienus Patemus, “a most zealous champion of military law”, cf 1.8.

‘’Cf R.W. Davies, “Joining the Roman army”. Service in the Roman Army, ch.l, 3-30.
He saw probatio as followed by the recruit’s arrival at his unit with a letter from the
provincial governor and a viaticum, followed in turn by four months’ training and tests
of proficiency before signatio and formal oath-taking and passing-out parade as a fully
qualified and trained soldier (p-26); however he was building on V. Cf 1.8.

BOOK 11: THE ANCIENT LEGION

the “sacraments” of military service.' They swear by God, Christ and
the Holy Spirit, and by the Majesty of the Emperor which second to
God is to be loved and worshipped by the human race. For since the
Emperor has received the name of the “August”," faithful devotion
should be given, unceasing homage paid him as if to a present and
corporeal deity. For it is God whom a private citizen or a soldier serves,
when he faithfully loves him who reigns by God’s authority. The
soldiers swear that they will strenuously do all that the Emperor may
command, will never desert the service, nor refuse to die for the Roman
State.^

6. How many cohorts there are in a legion, and how many soldiers
in a cohort.

Note that in a legion there ought to be ten cohorts. But the First cohort
exceeds the remainder in the number of soldiers and in rank, for it seeks
out the most select men as regards birth and instruction in letters. It
protects the eagle, which was always the especial and distinctive sign in
the Roman army of a whole legion. It undertakes the worship of the
images of the Emperors, that is the divine and propitious signa!^ It has
1,105 infantry, 132 cavalry cuirassiers, and is called a milliary cohort.
This is the head of the legion; from it when there is to be a battle the
First line begins being drawn up. The Second cohort has 555 infantry.

'sacramenta. Cf. F.J. DOlger, Antike und Christentum II (1930) 268-280, The whole
process of oath-swearing, tattooing and enrollment was part of a religious dedication. V.
thinks of it as something akin to Christian rites such as Ash Wednesday or Baptism, in
which the candidates are marked (with the sign of the Cross). The contemporary word for
this was signo, consigno, cf. M. Bonnet, Le Latin de Gregoire de Tours (Paris 1890) 240.
^Augustus.

^V. probably serves the Christian theocratic State enforced by Theodosius 1 in seeking to
justify swearing by God and the Emperor, instead of traditionally by the Emperor alone,
and to persuade people who were in any case loyal to the Emperor that they were really
loyal to God. The original pagan oath was first administered to the expanded army of 216
BC, cf Livy 22.38.1-5. But V.’s appears to be the late-Roman form, cf hints in Zos.
4.26.1, 4.33.3, Amm. 21.5.10, 16.10.40, Nov. Th. 11 7.1 (439). It does not follow the
pattern of oaths sworn to uphold the salus of the pagan Emperors and their families, cf
T. Mommsen, Romisches Staaisrecht (Leipzig I887‘) 1.622 ft'., 11.792.

Vffigies on standards; also - standards themselves. V.’s reliable information on the First
cohort ends here. Cf Ps.-Hyginus 3. The plan of Inchtuthil shows that it had 10 normal¬
sized centuries and only 5 centurions, the primuspilus thus commanding a double century,
and the other 4 primi ordines commanding double centuries, cf L.F. Pitts, J.K. St. Joseph,
Inchtuthil: the Roman legionary fortress excavations, 1952-65, Britannia monograph
series VI (London 1985) 164-169. The pilus posterior was apparently removed from the
First cohort before extant records begin.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

66 cavalry, and is called a quingenary cohort. The Third cohort
similarly has 555 infantry, 66 cavalry, but it is customary to approve
stronger men for this Third cohort because it stands in the centre of the
line. The Fourth cohort has 555 infantry, 66 cavalry. The Fifth cohort
has 555 infantry, 66 cavalry, but the Fifth cohort also needs strong
soldiers because like the First on the right the Fifth is placed on the left
wing. These five cohorts are drawn up in the First line. The Sixth
cohort has 555 infantry, 66 cavalry; to it, too, recruits of proven ability
should be assigned because the Sixth cohort stands in the Second line
behind the eagle and images. The Seventh cohort has 555 infantry, 66
cavalry. The Eighth cohort has 555 infantry, 66 cavalry, but it too needs
brave men because it stands in the centre of the Second line. The Ninth
cohort has 555 infantry, 66 cavalry. The Tenth cohort has 555 infantry,
66 cavalry; it also is accustomed to take good warriors because it holds
the left wing in the Second line. With these ten cohorts the full legion
is formed, having 6,100 infantry, 726 cavalry.' Therefore there should
not be a smaller number of soldiers in a legion; but sometimes a greater
number is customary if it shall have been ordered to receive more than
one milliary cohort.^

7. The titles and grades of officers of the legion.

Having set out the ancient battle-order of the legion, I shall list the titles
£ind ranks of principales milites or, to use the proper term, principia^
according to present-day rolls.^ The greater tribune is appointed by

‘So the “vulgate”, which gives the correct sum of cavalry; the best MSS offer 730 or 736.
The sums are artificially conceived from the basic assumption of (100 contubernales x 5
centuriae =) 500 contubernales + (10 decani x 5 centuriae =) 50 decani + 5 centuriones
— 555 infantry and (32 turmales x 2 turmae =) 64 turmales + 2 decuriones = 66 cavalry
for cohorts II-X, doubled for cohort I.

^V. conceives the legion as made up of combinations of milliary and quingenary cohorts,
when the terms otherwise refer exclusively to auxiliary units of the Principate. His
presupposition may be determined by late-Roman unit-sizes. The standard hypothesis of
the size of the “new legion” created by Diocletian (?) is 1,000 men, cf. E. Ritterling, RE
XII (1950) 1350, Hoffmann, 1.4, 215.

^V. confuses principia with principales: whereas the latter properly means “under¬
officers”, the former means “officers” in general, cf Amm. 22.3.2: praesentibus
lovianorum Herculianorumque principiis et tribunis, “in the presence of the principia and
tribunes of the Joviani and die Herculiani”, 25.5.1: collecti duces exercitus advocatisque
legionum principiis et turmarum super creando principe consultabant, “the assembled
generals of the army summoned the principia of the legions and cavalry troops and
consulted about creating a new Emperor”. 28.6.17: numerorum principiis, “the principia
of the units”, 15.5.16: principiorum vertices —“top brass” or senior army officers under
Silvanus magister peditum in Gaul ad 355. V. appears to list such officers of the “ancient

BOOK II: THE ANCIENT LEGION

sacred letter on the Emperor’s judgement. The lesser tribune arrives by
actual work.' The tribune is named after the tribe, because he is in
charge of soldiers whom Romulus originally chose by tribe.^ OrdinariP
is the title of those who, since they are first,'* lead the ordines^ in battle.
Augustales denotes those who were added to the ordinarii by Augustus.
Flaviales also, as second Augustales, were added to the legions by the
deified Vespasian.^ Eagle-bearers’ carry the eagle. Image-bearers* carry
the images of the Emperor. Optiones^ are called after “adopting”,'®
because when the above officers are out of action through illness, these
are accustomed to take charge of everything as their adopted
substitutes." Standard-bearers'’ carry the standards; they are now called

legion” as still survived in the modem organization, before going on to describe the
obsolete officers in II.8-14. The modem list seems to be related in the order of items to
the much longer list of “parts of the legion” given by loh. Lydus de Mag. 1.46.

'i.e. in the unit. The distinction probably refers to that between the late-Roman legiont^
commander (thhunus) and his vicahus, who might also call himself tribunus with
reference to function if he stood in for the tribune, although he was normally the
phmicerius of the unit Cf. Jones, 675.

^Cf. Varro de ling. lat. 5.81: tribuni militum, quod terni tribus tribubus Ramnium,
Lucerum, Titium olim ad exercitum mittebantur. “‘Tribunes’ of the soldiers, (so called)
because three each were sent by the three ‘tribes’ of the Ramnes, Luceres and Titles to
the army long ago.”

' = centurions.

*primi, perhaps like priores meaning “officers”.

^ = centuries. One of V.’s home-made etymologies, probably, like the two following.

^’On ordinarii, cf. J.R. Rea, “Ordinatus”, ZPE xxxviii (1980)217-219, J.F. Gilliam, “The
ordinarii and ordinati of the Roman Army”, Mavors II (Amsterdam 1986) 1-22; the term
ordinarius or ordinatus supplanted centurio in the late third century. Confusingly, it could
also refer to ordinary soldiers in an ordo or century, cf SHA Quadr. Tyr. 14.2. We know
that the adiutor counted among the officer ordinarii of the late-Roman army. MAMA
1.168: nau7x)u died Kap.m6ouKTdpcov <ap5evapiov), ibid. I.l69a: Eddv8po<;

adyouoTd Xioq dpi[0p]ou T(5vyEvv80TdT[(o]v Aayxid piq (fourth or fifth century, field-
army, Laodicea Combusta); P.Monac 1. no.8.46 (sixth century): 6p5(ivdpio<;)
dyou 0 xdXioq lefjvTi^; (limitanean, stationed at Syene); BGU 369 (sixth century):
(|)XaouiaA.fcp dpiGpou (t<»v Ka0oai(op(E'vQ)v) TpavJaTiypixavo) v {leg. pseud, cf
Hoffmann, 1.419^20), SB 1.4779 corrected by J.G. Keenan, “An instance of the military
grade Flavialis”, BASF x (1973) 43-46, q.v. (late-fourth century onward). Flaviales were
named after the Constantinian dynasty, not Vespasian, cf Keenan, loc. cit.

^aquiliferi.

^imaginarii. Otherwise always imaginiferi, which may be the true reading here too,
supported by several MSS. Cf R.W. Davies, “Some notes on military papyri from Dura”,
BASF V (1968) 31.

’centurions’ deputies.

^'^adoptandum. Cf Varro de ling. lat. 5.91, Paul. Fest. 201.23 L., for the etymology. In
the late Empire they were commissary officials and paymasters, cf Jones, 626, 667.

“vicars”, i.e. lieutenants.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

dragon-bearers.* Tesserarii announce the tessera^ through the soldiers’
10-man sections.^ The tessera denotes the general’s order, by which the
army is mobilized for some task, or for battle/ Campigeni,^ that is,
antesignani, are so named because it is owing to their efforts and
toughness that the general type (genus) of drill is improved on the
training-field (campus).^ Surveyors’ go on ahead and choose the site for
a camp. Beneficiarii^ are so called because they are promoted by the
beneficium^ of tribunes.'® Librarii^^ are so called because they enter in
books'^ the accounts pertaining to the soldiers. Trumpeters,'^
homblowers'^' and buglers'^ normally launch the battle with the trumpet,
bronze horn or bugle. Armaturae duplares are those who get double
pay;'® simplares get single pay. Quartermasters'’ measure out the places

^^signiferi.

'draconarii. Cf. 11.20. V. assumes just ten standard-bearers (= dragon-bearers), one for
each dragon, the late-Roman cohort standard. There were no longer such things as
centuries or century vexilla, described as antique in 11.13.

"password or written orders.

^contubernia, 8-man sections in Ps.-Hyginus 1.

'‘Cf, Amm. 21.5.13; 23.2.2, Ps.-Hyginus 43, viva tessera\ O. Douch, an archive from a
limitanean unit in Egypt c. ad 375, has several instances of a tesserahus, nos. 12, 15, 30,
33, 41, 53.

^ = campidoctores, arms instructors, drillmasters.

^'campigeni is unattested elsewhere. The rather far-fetched etymology is probably V.’s.
Antesignanus appears to denote a standard-bearer or senior NCO in Amm. 19.6.12: horum
campidoctoribus ut fortium factorum antesignanis... armatas statuas... heart lusserat
imperator. “The Emperor ordered armed statues to be set up in honour of their
campidoctores as leaders of their brave deeds.”

^Kfetatores^ cf. mertsores below.

‘^adjutants.

’patronage.

“Not only of tribunes, another of V.’s confusions. They continue to occur as staff-officers
in the bureaucracies of civil governors throughout the fourth century ad: P.Oxy. xlix
(1982) 3480, P.Lips. 55. CTh 8.4.7 (361) treats beneficiarii as a grade of cohortales. On
beneficiariiconsularisoiiht Principate, cf. N.B. Rankov, Beneficiarii Consularis, D.Phil.
thesis (Oxford 1986), and on beneficiarii generally, cf. E. Schallmayer et al. (edd.), Der
romische Weihebezirk von Osterburken /. corpus der griechischen und lateinischen
Beneficiarier-Jnschriften des romischen Reiches (Stuttgart 1990).

"clerks.

'^libri.

^''tubicines.

^^cornicines.

^^bucinatores.

“’Cf 1.13, where it is the instructors who are so treated.
mensores

BOOK II: THE ANCIENT LEGION

in camp according to the square footage' for the soldiers to pitch their
tents, or else assign them billets in cities.^ Wearers of the torque on
double, and on single pay:^ the torque of solid gold was a prize of
valour, winners of which sometimes gained double pay, in addition to
praise/ Duplares, sesquiplares: duplares got double pay, sesquiplares
got pay-and>a-half/ Candidati duplares, candidati simplores.^ These are
the under-officers,^ who are vested with privileges. The rest are called
“service-men’V because they have to do services.'^

^podismus.

^The Codes make no distinction between mefatores and mensores, cf. CTh 7.8, DE
METATIS, ib. 7.8,4 (393), mensorum. Mensores occur as government quartermasters in
CTh 7.8.4 (393), and metatores as the same in Th. II Nov. 25, line 24 (444), cf. CTh
7,8.10 (413); 7.8.16 (435). Sixth-century Byzantine sources do attempt some distinction:
loh. Lydus de mag. 1.46: |iq vacDpec;, Ttpopeipai — pqxdTCDpcc;,
Mauricius^/ra/.: dvxiKEvaopag {antecessores)f\xo\ TtpCTtapd lopaq {praeparatores) —
pnvCTOpag (1.3.35; 1.9; 2.12). Cf Ps.-Hyginus 37, 46, and M. Lenoir (ed ), ^ 127.

^torquati duplares, torquati simplores.

■^The torque was a contemporary decoration: Zos. 4.40.8, cf 2, in which a unit of Gothic
comitaiensesf?) stationed outside Tomi c. AD 386 and attacked by limitanei(?) under
Gerontius appear to have worn the torques as the badge of rank of their unit. Cf Ambr.
de ob. Val. 68.9: torques autem insignia esse victoriae dubitari non potest, cum hi, qui
in bello fortiter fecerint, torquibus honorentur. “That the torque is the insignia of victory
cannot be doubted, since those who have done bravely in war are honoured with torques.”
Cf also SHA Prob. 5.1, perhaps reflecting the age in which it was written. Cf also M.P.
Speidel, “The Master of the Dragon Standards and the Golden Tore”, TAPhA cxv (1985)
286-287 = Roman Army Studies II, 390-395. Why V. regards it as obsolete is unclear,
the more so as he has it from “present-day rolls”, but cf 11.3 on corruption having
“usurped the rewards of valour”.

^Late-Latin forms corresponding to duplarius, sesquiplarius, in use in the second and third
centuries AD, themselves abbreviated forms of duplicarius, sesquiplicarius.

^'TLL s. v. candidati militares (238 C) wrongly takes V.'s summary of the list as the
explanation of candidati. There were imperial bodyguards associated with the scholae
palatinae called candidati, cf Jones, 613 n.ll, Grosse, 96-97, but V's are less exalted.
In the third century AD there were candidati legion is, who were often seconded to
governors’ staffs like beneftciarii, but lowlier, cf Ann.Epigr. 1917.57. They are likely to
be beneftciarii in waiting; hence V.’s are principals in waiting. They fall naturally at the
bottom of any list of under-officers, and may have seemed self-explanatory to V. Cf optio
candidatus in M.P. Speidel, “The career of a legionary”, TAPhA cii (1982) 209-214 =
Roman Army Studies II, 197-202.

^milites principales. V. again seems to equate modern principia or “officers” with
principales.

'^muniftces.

^munera, cf. 11.19. Cf munifex appointed catafractarius in the same unit, J.R. Rea, “A
Cavalryman’s Career, A.D. 382 (?)-40r’, ZPE Ixvi (1984) 82.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

8. The titles of the officers who led the ancient centuries/

Ancient practice stipulated that the chief centurion^ was promoted from
the First centurion of the principes^ of the legion; he was in charge of
the eagle and controlled four centuries, that is, 400 soldiers, in the First
line. As head of the entire legion, he received benefits and perquisites,"*
Correspondingly, the First centurion of the hastatv' led two centuries,
200 men, in the Second line; he is now called a ducenarius!" But the
princeps of the First cohort^ controlled a century-and-a-half, 150 men.
Almost all general administration within the legion fell to him. The
Second centurion of the hastati^ also commanded a century-and-a-half,
150 men. The triariusprior^ controlled 100 men. Thus the ten centuries
of the First cohort were controlled by five centurions.'^ The ancients set
up great privileges and great honour for them, to encourage the rest of
the soldiers from the entire legion to strive with every effort and loyal
service to gain these great rewards.

'ordines. In contrast to the modem officers listed in 11.7 we have obsolete officers from
here to the end of 11.14, and then the old battle-array (11.15-17), followed by an appeal
to the Emperor to restore it (11.18), and the administrative arrangements of the “ancient
legion” (II. 19-end), with various hopeful modern equivalents (often incorrectly) supplied.

^centurio primi pili, centurion of the primus pilus.

^primus princeps^ i.e. primus princeps prior^ according to traditional terminology. Cf.
M.P. Speidel, “The Centurions’ Titles”, Epigraphische Studien XIII (Cologne 1983)
43-61, repr. with addendum in id., Roman Army Studies II, 21-39, for the correct forms,
which V. misconceives.

^commoda, cf. CV 12.62.1 (AD 253-268): commoda primipilatus post administrationem
deberi incipiunt, et si is, qui ea percipere debuit, prius rebus human is eximatur, heredibus
petitio salva sit. “Perquisites of the primipilate start to fall due after the period of service,
and if one who was due to receive them shall be removed first from human affairs, let the
claim be safe for the heirs.” According to B. Dobson in D.J. Breeze, B. Dobson, Mavors
X (Stuttgart 1993) 183, the commoda included a gratuity which bestowed equestrian rank.
Cf also M.P. Speidel, Roman Army Studies II, 368, and cf 11.21.

^primus hastatus, i.e. primus hastatus prior.

^’Probably V.’s equation, by a false etymology, ducenarius < ducenti. The ducenarius and
centenarius were middle-ranking late-Roman NCOs in vexillationes, auxilia and scholae,
but their functions are not clearly established, cf Grosse, 117-118, Jones, 634. V. likens
centurions to them.

^V. means again the First centurion of the principes.

'^secundus hastatus, i.e. primus hastatus posterior. V.’s terminology presupposes no more
than one cohort for the unit, when in fact he is trying to describe the structure of a ten-
cohort unit.

‘^ot a traditional title of a centurion under the Principate, when triarii were called pili,
and we have already met the primus pilus, who was not balanced by a pilus posterior in
the First cohort; in any case V. does not make the connexion. He has omitted the primus
princeps posterior.

or dinar a.

BOOK II: THE ANCIENT LEGION

There were also centurions, who looked after single centuries; these
are now called centenarii} There were “deans”^ too, in charge of ten
soldiers each; they are now called caput contubernii.^ The Second cohort
had five centurions, and so did the Third and Fourth, and so on to the
Tenth, cohorts. In the whole legion there were 55 centurions.^*

9. The duties of the Prefect of the legion/

Ex-consuls used to be sent out to the armies as legates of the Emperor.^
All legions and auxilia obeyed them in the administration of peace or
the exigency of war. In their place the Illustrious Men^ the Masters of
the Soldiers* are, it is well-known, now substituted, controlling not
merely pairs of legions, but greater numbers also. But the individual
commanding officer^ was the Prefect of the legion, who had the rank of
“Count of the First class’’.''^ In the absence of the legate, he retained as
his “vicar”'' supreme power.'^ The tribunes, centurions and the rest of
the soldiers obeyed'^ his commands. The tessera^"^ for the night-watches

‘Probably V.’s equation, by a false etymology, centenarius < centum See note above
under ducenarius.

^decani.

^“captain of 10-man section”. Decani do not appear in our records before the fourth
century AD. The caput contubernii may be identical with the caput z from Concordia
cemetery, cf. Hoffmann, 11.28 n.206; 1.76. The title seems only temporarily to have
displaced decanus.

‘‘i.e. 55 centuries, but V. forgets that the 10 centuries of the First cohort were under 5
ordinarii, so that his figures add up to 50 centurions only.

^praefectus legionis. The title is not known before the second century ad; cf. B. Dobson,
Die Primipilares (Cologne 1978) 69.

^’i.e. as consular governors. Cf C. Lessing, Scriptorum historiae Augustae Lexicon
(Leipzig 1901-06) 88, s.v. “consul 5)”.

^inlustres viri, the highest rank of the fourth century AD.

^magistri militum.

'^iudex or “judge”, a general term for provincial governor or equivalent.

'^'comitiva ordinis primi. See Ill.l for the equation of the ancient consul with the modern
magister militum, and the ancient praetor with the modern comes rei militaris. Here V.
thinks of the praefectus legionis as like a comes rei militaris. On the praef leg., identical
to the praef. castr., cf B. Dobson, loc. cit., 68-74.

^'vicarius, i.e. deputy.

'^This parallels the relation between the tribunus and vicarius = minor tribunus in 11.7.
V. does not appear to know of the tribunus laticlavius who ranked between the legate and
the praef. leg. before his abolition c. AD 260. Cf. M. Christol, Essai sur revolution des
Carrie res senatoriales dans la seconde moitie du I He siecle ap. JC (Paris 1986) 39-44.

'^On the late-Latin vogue for servo = observo, popular in V., cf. A. Andersson, 46.
‘'‘written order or password.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

or the march was obtained from him.' If a soldier committed some
crime, he was sent by a tribune for punishment on the authority of the
Prefect of the legion. The arms of all soldiers, likewise the horses,
uniforms and pay and rations^ were part of his responsibility. The
enforcement of discipline, and the training both of infantry and
legionary cavalry were looked after daily at his command. In his person
he was just, conscientious and sober, and by means of constant labours
he trained the legion entrusted to him in every act of loyal and attentive
service, in the knowledge that the excellence of those under him
redounded to the Prefect’s praise.

10. The duties of the Prefect of the camp.

There was also a Prefect of the camp,^ who while inferior in rank was
still occupied with important concerns. He was responsible for the siting
of the camp and the assessment of the rampart and fosse. The soldiers’
tents or huts, along with all baggage, were taken care of under his
command. Moreover sick soldiers, the doctors who tended them, and
payments,'* belonged to his duties. He saw to it that there were never
any shortages of wagons, pack-horses, or iron tools for sawing and
felling timber, opening fosses, building a rampart or an aqueduct;
likewise, of firewood, straw, battering-rams, mangonels,^ catapults^ and
other kinds of torsion-engines. He was selected after lengthy and tested
service as the most experienced of the men, to teach others correctly
what he had himself done with merit.’

'The praef. leg. had a tesserarius on his staff, cf. M.P. Speidel, Roman Army Studies I
(1984) 287.

^annona.

^The praefectus castrorum was a creation of Augustus, cf. Dobson, loc. cit., 68. The
division of duties between this and the preceding officer is probably invention; the two
are not attested simultaneously in the same legion. The duties also are perhaps invented
along etymological lines. What we know of the duties of the praef. castr./leg. implies that
he was a campaigning regimental commander, not a bureaucrat tied to the camp, cf
Dobson, 71, and Tac. Ann. 12.38.3, 13.39.2.

^expensae, cf III.3.

^onagri, or stone-throwing engines.

^'ballistae, or arrow-shooting machines.

^At 111.2 the praefectus legionis and tribuni and principia are all responsible for the care
of the sick; the annona is taken care of by the praefectus legionis in 11,9, but expensae
belong to the praefectus castrorum in 11.10; the training of the men is the duty of the
praefectus legionis in II.9, but teaching responsibility belongs to the praefectus castrorum
in II. 10. Also the duty to see that supplies are constantly available appears to be arbitrarily
divided between the praefectus castrorum in II. 10 and praefectus fabrum in II.l 1.

BOOK II: THE ANCIENT LEGION

11. The duties of the Prefect of engineers.'

Moreover the legion has engineers, carpenters, masons, wagon-makers,
blacksmiths, painters and other artificers, ready-prepared to construct
buildings for a winter camp, or siege-engines, wooden towers and other
devices for storming enemy cities or defending our own, to fabricate
new arms, wagons and the other kinds of torsion-engines, or repair them
when damaged. They used to have workshops, too, for shields, cuirasses
and bows, in which arrows, missiles, helmets and arms of every type
were made.^ For the main aim was to ensure that nothing which the
army was thought to require should be lacking in camp, to the extent
that they even had sappers^ who, after the fashion of the Bessi,"' sank
mines underground, and dug through the foundations of walls to emerge
unexpectedly and capture hostile cities.^ The particular officer
responsible for these matters was the Prefect of engineers.^

12. The duties of the Tribune of soldiers.’

We have said that a legion has ten cohorts. But the First was a milliary
cohort, into which were sent soldiers favoured by their wealth, birth,
literacy, physique and strength. This was under the command of the

'praefectus fabrorum. V. uses the unusual form of the gen. pi. here and perhaps,
according to some of the best MSS, below. This officer was an aide-de-camp, often only
in an honorary capacity, in the late Republic and early Empire, so the duties assigned to
him here are probably an etymological interpretation. It may be argued from Paternus
apud Dig. 50.6.7 that the optio fabricae and architectus took care of the technical side of
the legion’s activities and services under the Principate. See B. Dobson, “The Praefectus
Fabrum in the Early Principate”, in D.J. Breeze, B. Dobson, Mavors X (Stuttgart 1993)
218-241.

^Under Diocletian’s reforms these services were centralized in permanent munitions
factories, cf. S. James, “The fabricae: State arms factories of the Later Roman Empire”,
in Roman Military Equipment Military Equipment and the Identity of Roman Soldiers.,
Procs. of the 4th. Roman Military Equipment Conference, ed. J.C. Coulston, BAR Int.
Ser. 394 (Oxford 1988) 257-332.

^cunicularii.

‘’Cf IV.24. The Bessi were a gold-mining tribe in Thrace, cf Claud, de cons. Mall.
Theod. (AD 399) 38-41; Pac. Pan Lat. 2 (12).28.2; cf Amm. 31.6.6.

^Ordinary legiones palatinae and auxilia palatina dug mines, cf Amm. 24.4.13,
24.4.21-22 called cuniculariU Zos. 3.22.2. Perhaps V.'s cunicularii were the technical
consultants for the operations; more likely, however, V. is enlarging upon jejune source-
material off the top of his head.

^'praefectus fabrum. D. Baatz, “KatapultbewatTnung und -produktion des rOmischen
Heeres in der frUhen und mittleren Kaiserzeit”, Mavors XI (Stuttgart 1994) 127-135, at
133, argues that V. has substituted praef fabr. for praef. castr. in his source because of
the etymology.

^tribunus militum.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

tribune who was outstanding in military knowledge, physical strength
and honourable character. The remaining cohorts were governed by
tribunes or “provosts”,* according as the princeps decided.'

The supervision of training the soldiers was taken very seriously.
Tribunes or “provosts” would order the ordinary soldiers entrusted to
them to train daily under their gaze, but also encouraged them to imitate
their own example, since they were themselves perfect in the art of
bearing arms. It is the tribune who is praised for his conscientiousness
and hard work, when a soldier goes forth immaculate in his uniform,
well-protected and gleaming in his arms, and versed in the practice and
system of his training.^

13. The centuries and ensigns'* of infantry.

The chief standard^ of the entire legion is the eagle, carried by the
eagle-bearer.^ Dragons are also carried into battle one for each cohort
by dragon-bearers.^ But the ancients knew that in battle once fighting
commenced the ranks and lines quickly became disordered and
confused. To avert this possibility they divided the cohorts into centuries
and established individual ensigns for each century. The ensign was
inscribed with letters indicating the century’s cohort and ordinal number
within it. Seeing and reading this, soldiers could not stray from their

'praepositi, i.e. “officers commanding”.

^princeps, i.e. the senior administrative officer equated with the old First centurion of the
principes (11.8). V.’s idea of him is anachronistic. As the chief of the bureaux of all
militaiy officials, including the magistri militum, comites and duces, the late-Roman
princeps wielded great power; cf Grosse, 120 ff. V.’s conception of tribunes resembles
the major and minor types of the modern army (see 11.8) rather than the
laticlave/angusticlave of the Principate. The minor type appear interchangeable with
praepositi also in C/ 12.42.1 (323): ne cui liceatpraepositorum vel tribunorum cohortium
vel vicariorum... R.E. Smith, “Dux, praepositus”, ZPE xxxvi (1979) 263-278, traces the
histoiy of the term praepositus without always separating legionary vexillations from
auxiliary. Cf M. Christol, “L’arm^e des provinces pannoniennes et la pacification des
revokes maures sous Antonin le Pieux”, Ant.Afr. xvii (1981) 133-141 on auxiliary
vexillations.

^The trainIng-r6le of tribunes is identical to that already given to the praefectus legionis
in II.9. Hence the details are probably reconstructed.

\exilla.

^signum.

^'aqui lifer.

^draconarii.

BOOK II: THE ANCIENT LEGION

comrades, whatever the confusion of battle.* They also detailed
centurions, now called centenarii —very much battle-ready, clad in a
cuirass,^ wearing transverse crests on their helmets to make it easier for
them to be recognized—to command individual centuries. This was so
that no deviation would arise, since the soldiers in groups of a hundred
followed their own ensign and the centurion who had the sign on his
helmet. The centuries were themselves subdivided into lO-man sections.^
For every ten soldiers living under one tent, there was one in charge as
“dean”,^ now called “section-captain”.^ The 10-man section used to be
called a “maniple”^ because they fought in groups {mantis) joined
together.^

14. The troops* of legionary cavalry.

As the term “century” or “maniple” is used among infantry, so the
corresponding expression among cavalry is the turma. One troop
contains 32 cavalrymen. The officer-in-charge is called a“decurion”. As

'V. forgets that the average soldier could not read, as we know from numerous documents
on papyri written by amanuenses and as we can see from 11.19. The description of the
vexilla is probably V.’s own reconstruction, prompted by information in his source about
shield-emblems and the jabelling of the shield with the owner’s name and century, whose
purpose (to prevent mistaken ownership or theft) V. has wrongly lumped together with
that of the emblems, cf II. 18. The existence of century standards under the Principate
is uncertain, cf W. Zwikker, “Bemerkungen zu den rOmischen Heeresfahnen in deralteren
Kaiserzeit”, BRGK xxvii (1937) 7-22. It seems certain that they did not exist under the
Republic, cf Varro de ling. lat. 5.88, Livy, 27.14.8,

M.e. not bureaucrats as the few remaining centurions are likely to have been in the fourth
century AD. Cf also 11.8.

^contubernia.

*decanus.

^caput contubernii, cf 11.8.

^'manipulus.

^V. seems to regard manipulus as a diminutive of manus. Cf Ovid Fasti 3.118 for the
conventional identification with the “handful” of straw of the primitive standard. See also
R. Maltby, A lexicon of ancient Latin etymologies (Leeds 1991), s.v. manipulus. Varro de
ling. lat. 5.88, cf 6.85, derived it from mantis —groups. For the late-antique use of
manipulus = contubernium., cf SHA Pesc. Nig. 10.5-6, of a unit of 10 or 11
commanipulones. For the size, cf also Maur. Strat. 12. B (8) 9, 12. B (8) 6: Katd
bempxiav fjtoi kovtouPe' pviv “by decarchy or contubernium." In Ps.-Hyginus Castr. 1
the tent covers only an 8-nian mess,

*turmae.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

110 infantrymen are controlled by one centurion under one ensign, so
32 cavalrymen are governed by one decurion under one ensign.’

Moreover, in the same way as a centurion is chosen for great strength
and tall stature, as a man who hurls spears and javelins skilfully and
strongly, has expert knowledge how to fight with the sword and rotate
his shield and has learned the whole art of armatura,^ is alert, sober and
agile, and more ready to do the things ordered of him than speak, keeps
his soldiers^ in training, makes them practise their arms, and sees that
they are well clothed and shod, and that the arms of all are burnished
and bright,—in the same way the man who is to be chosen as decurion
to be put in charge of a troop of cavalry, should above all be physically
able to mount a horse while cuirassed and girded with all his arms in
highly impressive style, to ride heroically, wield the lance" with skill
and shoot arrows expertly, instruct his turmales, that is, cavalrymen
placed under his charge, in all things needed for cavalry warfare, and
make them frequently clean and look after their cuirasses and
cataphracts, lances and helmets. The glitter of arms strikes very great
fear in the enemy,^ Who can believe a soldier warlike, when his
inattention has fouled his arms with mould and rust? It is advantageous
to school not Just the men, but the horses too through constant training.
So responsibility for the fitness and training of both men and horses
devolves upon the decurion.

'There were no decuriones in the legion of the Principate, cf. M.P. Speidel, Roman Army
Studies I (1984) 294, id. “The captor of Decebalus: a new inscription from Philippi”, JRS
Ix (1970) 145 n.34, D. Breeze, “The organization of the legion: the First cohort and the
equites legion is”, JRS lix (1969) 50-55. The source derives from Cato, in whose day there
were, cf. Polyb. 6.25.1—2, describing 10 turmae (30-strong) each under 3 decuriones, one
of whom was senior commander of the turma. Cf Livy 4.38.2, 22.38.3-4.

^special drill, cf 1.13.

^contubernales,

^contus.

^e.g. Gratian’s court troops in 378, Amm. 31.10.10, 14. Cf Donatus ad Verg. Aen. 7.626:
quoniam Juigor armorum plurimum hostibus terror is importat, “since the shine of arms
conveys very great fear to the enemy”, and 8.402: plurimum enim terrorem hostibus
incutit etiam splendor armorum, “for the glitter of arms also strikes very great fear in the
enemy”, with Vegetius’ words here: Plurimum enim terroris hostibus armorum splendor
inportat.

BOOK II: THE ANCIENT LEGION

(The Ancient Legion in battle-order, ch. 15-18)

15. How the lines of the legions are drawn up.’

Next, the manner in which the battle-array should be drawn up for
imminent battle may be demonstrated from the example of one legion.
This could be extended to more if need demands.

The cavalry are stationed on the wings. The infantry line begins to be
ordered from the First cohort on the right wing. It is joined by the
Second cohort. The Third cohort is placed in the middle of the line. The
Fourth is placed next to it. The Fifth cohort holds the left wing. Those
fighting before the standards, around the standards and (otherwise) in
the front line were called principes^ [i.e. the centurions^ and the other
under-officers].'* This was the heavy armament, which had helmets,
cataphracts, greaves, shields, large swords called spathae, and other
smaller swords called semispathia,^ five lead-weighted darts'^ slotted
inside their shields, to be thrown at the first assault, and also two
Javelins, one of which was larger, with a triangular iron tip 9 in. long,
and a shaft of S'A ft.; it used to be called a pilum, and is now known as
a spiculum. Soldiers were particularly trained at throwing this type,
because when aimed with skill and force it often transfixes shield¬
bearing infantry^ and heavy cavalry cuirassiers.^ The other was smaller,
with an iron tip 5 in. long and a shaft of VA ft.; it was once called a
vericulum, and is now known as a verutum!^

'Essentially the same archaic manner as at 1.20, except that there the light-armed are on
the wings, and as at III. 14, except that there the light lines are counted separately. V. also
substitutes late-antique soldiers and weapons, creating anomalies.

^“frontlinesmen”.

^ordinarii.

^principales. The bracketed text may be a marginal gloss that has become interpolated.
V. or a scholiast mistakes principes for phncipia or principales Cf. 11.7 for principia =
principales.

The spatha and semispathium are late-Roman weapons substituted for the gladius and
pugio. Cf. Isid. Etym. 18.6.4-5. The spatha or broadsword was an auxiliary weapon in the
first century AD, cf Tac. Ann. 12.35.

^'plumbatae, cf 1.17.

^scutati pedites.

^ioricati equites.

‘The term vericulum or “little spit” is not otherwise attested as a weapon The pilum and
vericulum described are likely to be only nominally archaic. The pilum at 1.20 had a head
measuring 9 in. or 1 ft., and was also triangular. The special design “so that it could not
be broken off’ is paralleled by Polyb. 6.23.9, whereas later designs beginning with that
with one wooden pin introduced by C. Marius at Vercellae, 101 BC (Pint. Mar. 25.1) and

VEGETIUS’ EPITOME OF MILITARY SCIENCE

The second line was similarly armedJ The soldiers who took up
position in it were called hastati} In the second line, the Sixth cohort
was placed on the right wing. The Seventh joined it. The Eighth cohort
held the middle of the line, accompanied by the Ninth. The Tenth
cohort always held the left wing in the second line.

Such were the arms in which the first line of principes and the second
line of hastati are considered to have been arrayed. Behind them were
the light troops^ and light armament, whom we now call exculcatores"^
and armaturae,^ and scutatf equipped with lead-weighted darts, swords
and javelins, just as almost all soldiers seem to be armed todayThere
were also archers armed with helmets, cataphracts and swords, in

ones with variable hardness of steel described by Caes. Bell. Gall. 1.25, Arr. Ect. 17 ff.,
App. Celt. 1.1, were designed to bend at the joint upon impact. However the length of
V.’s heads is much smaller than sizes prescribed for the old javelins by Polyb., Dion. Hal.
5.46, Appian loc, cit. It seems unlikely that soldiers normally carried the weight of arms
prescribed by V.; cf. IV.44 for a similar weight given to marines, where it seems still
more inappropriate if they fought in loose order. Probably V. exaggerates to counteract
the alleged tendency to discard heavy armour, cf. 1.20.

'This and the following paragraph are here transposed, after A. Gemoll, “Exercitationes
Vegetianae”, Hermes vi (1871) 113-118. The distribution of cohorts is the same as at II.6.

^“spearmen”.

^ferentarii.

^skirmishers.

^Here and in 11.17 apparently not meaning those versed in the special drill called
armatura, but simply “light-armed”; cf. Amm. 14.6.17: deinde leves armaturas, post
iaculatores... Exculcatores {aux. pal. in Not. Dig. occ. 5.173 = 7.20 exculcatores seniores,
5.175 = 7.122 exculcatores iuniores, 5.207 exculcatores iuniores Britanniciani), known
as OKOuXKCt Topeq in Byzantine sources (Maur. Strat. 1.3.20: oKouXKd Tope(; oi
KaidaKOTTOi Aiyoviai, 2.10 (11): HEPl KATAIKOnON H ZKOYAKATOPON on
scouting during the battle, 4.5.1 where oKOoXKa is one of the activities for which the
drwwgwj-formation is best suited, 7. B 13: FIEPl SKOYAKAI on scouting before the battle,
7.17a.6, 9.5.), are associated with armaturae and Junditores in V. Epit. U. 15-17, and
clearly resemble Ammianus’ proculcatores in both function and word-formation (Amm.
27.10.10, where they scout out routes like excursatores, cf Amm. 24.1.2, also Maur.
Strat. 9.5.18,4.3.9: TipocrKouXKeif eiv);cf alsoTheophyl. 6.9, referring the term (TKouXKct
to the sentries of the watch. Etymologists are inclined to believe the word is of Germanic
rather than Latin origin, cf A. Walde, J.B. Hofmann, Lateinisches Etymologisches
Worterbuch (1938-54^) 502, s.v. sculca, A. Ernout, A. Meillet, rev. edn. J. Andr6,
Dictionnaire Etymologique de la langue latine (1979'') 606, s.v, sculca.

'^shield-bearing (light) infantry.

’Cf Procop. Bell 1.1.8-16. where we learn of early Byzantine mounted archers equipped
with shields attached to the shoulder, without encumbering the hand, cf Scutari! Sagittarii,
scholae palatinae, archers being a light-armed category which does not usually carry a
shield. There is no reason to think that such archers could only perform from horseback.
V.’s scutati also belong to both heavy and light lines. The modem equipment does not fit
the organization of the array, which implies more distinct equipment for the different
rdles.

BOOK 11: THE ANCIENT LEGION

addition to bows and arrows.‘ There were slingers who fired stones from
slings^ and “sling-staves”.^ There were tragularit who shot bolts with
catapults^ and crossbows.^

16. How triarii and centurions were armed.

Behind all the lines were posted the triarii,^ armed with shields,
cataphracts, helmets, greaves, broadswords,® semispathia^ lead-weighted
darts, and a pair of javelins. They waited in reserve on bended knee, so
that in the event of the front lines being defeated, the battle could be
restored by them as if anew, and victory retrieved.'^ All antesignani^^
and standard-bearers,'^ though infantry, received small cuirasses, and
(leather) helmets covered with bearskins to frighten the enemy.
Centurions, however, had cataphracts, shields and iron helmets with
transverse silvered crests for swifter recognition by their men.'^

17. When battle commences the heavy armament stands like a
wall.*^

This point should be noted and maintained by every manner of means:
when battle commenced, the First and Second lines stood immobile. The
triarii also sat inactive. Meanwhile the light troops,'^ armaturae.

'Cf. Procop. ibid.

^adfundas =Jundis, cf. 11.23 ad omne genus... armorum, ...ad palum, 111.24 ad latiores
lanceas, IV.8 ad Jundas, IV. 18 ad maiores ballistas, examples of late-Latin instrumental
ad.

^Justibali.

Vatapultiers and crossbowmen.

^manuballistae.

^'arcuballistae. Tragularii were probably archaic; V. is the sole source for these
specialists, but cf. Livy 24.42.2: Cn. Scipionisfemur tragula confixum erat (214 BC). “Cn.
Scipio’s thigh had been pierced by a tragula.'' Manuballistae and arcuballistae were
characteristic late>Roman weapons (cf IV.22). V. simply updates a manipular legionary
array derived from Cato de Re Militari.

’“thirdlinesmen”.

*gladn = spathae.

’^short swords or daggers.

“’Polyb. 6.23.16 and Livy 8.8.10 arm triarii with the hasta or thrusting spear, but V.
assimilates their arms to his plan for the principes and hastati.

" = campidoctores, cf II.7, or simply infantry NCOs, cf II.2.

'^signiferi.

"Cf 11.13.

"Cf the Homeric tactic, II. 4.299. Cf also 1.20.

^ferentarii.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

exculcatores, archers, and slingers, that is, the light armament,'
provoked the opposition, going in front of the line. If they managed to
put the enemy to flight, they pursued. If they came under pressure from
the other side’s resolve or numbers, they returned to their own men and
took up position behind them.^

Then the heavy armament^ took up the battle, and stood so to speak
like a wall of iron, fighting it out with javelins and at close quarters
with swords. Even if they routed the enemy, the heavy armament did
not pursue, lest they disturb their own line and battle-order, and the
enemy charge back on them while dispersed and overwhelm them when
disordered, but the light armament, with slingers, archers, and cavalry,
pursued the fleeing foe. By adopting this disposition and these
precautions, the legion would win without incurring danger, or if
overcome was preserved intact. For the rule of the legion is neither to
flee nor pursue easily.

18. The names and ranks of soldiers are to be written on the face
of their shields.

To prevent soldiers straying from their comrades at any time in the
confusion of battle, they painted different signs for different cohorts on
their shields, digmata^ as they call them themselves, and it is customary
to do this even now.^ Also the name of each soldier was inscribed in
letters on the face of his shield, with a note of which cohort or century
he was from.^

It is clear from the above, then, that a well-trained legion was like a
very well-fortified city, which carried all that was essential for battle

' levis armatura. Cf. 11.15.

'Cf Onas. 19.

^gravis armatura.

'’Greek for “designs”.

^On the designs in the Notitia Dignitatum^ cf. R. Grigg, “Inconsistency and Lassitude: the
shield emblems of the Notitia Dignitatum^\ JRS IwVu (1983) 132-142, M.P. Speidel, “The
Army at Aquileia, the Moesiaci Legion, and the Shield Emblems in the Notitia
Dignitatum'\ SJ xiv (1990) 68—82, = Roman Army Studies II, 414-418.

^’Cf Cass. Dio 67.10.1 on the introduction of the custom of inscribing the soldier’s name
and the name of his centurion on the shield in c. AD 89 by Tettius Julianus, cos. 83. V.
customarily fails to distinguish between cohorts and centuries, probably because there was
no distinction observed by the small legions of his day.

BOOK II: THE ANCIENT LEGION

around with it everywhere, and feared no sudden hostile assault:* even
in the midst of plains it would fortify itself upon an instant with fosse
and rampart; it contained within itself all types of soldier and arms. If
anyone, therefore, desires the defeat of the barbarians in an open battle,
let him seek in all his prayers, that by the will of God and the
Invincible Emperor’s policies, the legions may be reinstated with new
recruits. Within a brief space of time, recruits carefully selected and
trained every day, not just in the morning but even in the afternoon, in
every skill of arms and art of warfare, will easily match those soldiers
of old who conquered the entire terrestrial sphere. Let it not be a
problem that customs which were thriving have long since changed.
Your Eternity’s good fortune and foresight are such as both to devise
innovations for the safety of the State, and to restore ancient principles.
Every work appears difficult before you attempt it, but if trained and
careful men are put in charge of the levy, a company fit for war can
soon be assembled and thoroughly drilled. Skills can achieve anything
you wish, if adequate funding is not spared.^

(The administration of the Ancient Legion, ch. 19-25)

19. As well as physical strength, skills in short>hand writing and
computation are required of recruits.

Since there are several administrative departments^ in the legions which
require literate soldiers, it is advisable that those approving recruits
should test for tall stature, physical strength and alertness in everyone
indeed, but in some the knowledge of “symbols”'* and expertise in
calculation and reckoning is selected. For the administration of the
entire legion, including special services,^ military services,^ and money.

'The image of a city on the move is influenced by the wagon-city or carrago of the
Goths, Huns and Alans in the late-fourth century ad. At III. 10 fin. V. recommends
imitation of the carrago.

^In this epilogue to the “ancient legion” in battle-order V. commends it as a model of
self-sufficiency, training and tactical flexibility, in implied contrast to the contemporary
field armies. The model is conceived by V. in terms of a late-Roman field army as it
could become, not as an essay in antiquarian reconstruction.

^scholae = qfficia or scrinia, i.e. offices.

^notae, short-hand writing.

^obsequia. Limitanei might be lent on secondment to the staff of ducal commanders as
Serjeants and orderlies; typical duties involved acting as messenger, porter, and prison
warder. Comitatenses might be seconded to the staff of magistri miliwm. Cf Jones,
597-598.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

is recorded daily in the Acts with one might say greater exactitude than
records of military and civil taxation are noted down in official files.’

Daily even in peacetime, soldiers take it in turns from all centuries
and 10-man sections^ to do night-watch duties,^ sentry duty,"* and
outpost-duties.^ The names of those who have done their turn are
entered in lists so that no one is unjustly overburdened or given
exemption. When anyone receives leave of absence^ and for how many
days, it is noted down in lists. For in antiquity it was difficult to be
given leave unless for very good approved reasons. Established soldiers
were not seconded to any special services at all, nor were they
employed for private business. It seemed incongruous that a soldier of
the Emperor, maintained in uniform and pay and rations’ at public
expense, should have time to serve private interests.^ Instead, soldiers

^'munera, ordinary officers’ s«^rvices, water, fodder and firewood; see below, and cf also

'G.R. Watson, “Documentation in the Roman Army”, AI^RW II. 1 (1974) 493-507, R.O.
Fink, Roman Military Records on Papyrus (Cleveland, Ohio, 1971) A.K. Bowman, J.D.
Thomas, The Vindolanda Writing-Tablets, Tabulae Vindolandenses II (London 1994); cf
SHA Alex. Sev. 21, Rufinus adv. Hieron. 2.36, Isid. Etym. 1.24.1.

^contubernia.

\igiliae.

* excub itus.

^agrariae.

^commeatus. The corrupt granting of leave was a standing abuse in the army of the late
fourth century, cf Jones, 633, 639, 644-645, 648-649, Grosse, 246-248. See also M.P.
Speidel, “Furlough in the Roman army”, YClSxwlu (1985) 283-293, = id., Roman Army
Studies II (Stuttgart 1992) 330-341.

^annona.

*An abuse banned by Augustus but always practised; cf Patemus apud Aem. Macer de
Re Mil. I {Dig. 49.16.12 §1): Officium regentis exercitum non tan turn in danda sed etiam
in observanda disciplina constitit. Paternus quoque scrips it, debere eum, qui se me miner it
armato praeesse, parcissime commeatum dare, equum militarem extra provinciam duci
non permittere, ad opus privatum piscatum venatum militem non mittere, nam in disciplina
Augusti ita cavetur: '‘Etsi scio fabrilibus operibus exerceri milites non esse alienum,
vereor tamen, si quicquam permisero, quod in usum meum aut tuum fiat, ne modus in ea
re non adhibeatur, qui mihi sit tolerandus ”. “The duty of an army general consists not
only in enforcing discipline but also in observing it. Paternus also wrote that a
conscientious leader of soldiers should grant leave very sparingly, should not permit a
military horse to be taken outside the province, should not detail soldiers for private
business, fishing or hunting, for in the Discipline of Augustus it is laid down; 'Although
I am aware that it is not inappropriate for soldiers to be exercised in building operations,
yet I am afraid that if I permit anything which may be done for my or your benefit, a due
proportion may not be observed in that activity, which would be acceptable to me’.” CJ
12.35.13 (AD 398) orders dismissal for soldiers caught pursuing their own or another’s
private interests {vel sibi vacet vel aliena obsequia) without permission of their
commanding officer {sine nutu principali). Those found keeping soldiers for private
purposes were fined 51b. gold. Soldiers sent as couriers were to execute their commissions

BOOK II: THE ANCIENT LEGION

called “extras’” were assigned to serve commanders,‘ tribunes and
under-officers,^—so called because they had been added after the legion
had been filled."* They are now called “supernumeraries”.^ However,
even established soldiers would carry “bundles”,^ that is, firewood, hay,
water and straw, into camp. They are called “service-men”^ because they
perform these services.*

directly and promptly or face arrest for being absent without leave. CJ 12.35.15 (ad 458)
orders fines for soldiers caught pursuing private business and for their employers, as
soldiers were supposed to be training daily (sedprophae muniis insudare militiae... sed
frequentes esse in numero suo iubeat, ut armorum quotidiano exercitio ad beiia se
praeparent),

^accensi.

^ indices.

^principales.

^Cf. accensi in the Catonian battle-order at V. Epit. 111.14, and the Servian “constitution’',
Livy 1.43, and Livy 8.8.10, Paul, Fest. 13.23 L. s.v. adscripticii, Festus 216.23 L.: Optio
qui nunc dicitur, an tea appellabatur accensus. Is adiutor dabatur centurioni a tribune
militum..., “the optio as he is now called was once termed accensus. He was given to a
centurion as an assistant by the tribune of the soldiers...”; ibid. 506.26 L.: s.v. velati,
...Cato eos ferentarios dixit, qui tela ac potiones militibus proeliantibus ministrabant,
“Cato called them ferentarii, who served weapons and drinks to the fighting soldiers”;
Varro de ling. lat. 7.58; Accensos ministratores Cato esse scribit (Jordan p.81, Cato de
Re Mil. fr. 8), “Cato writes that accensi are servants.”

^supernumerarii. Attested in the late-Roman civil service in this technical sense of men
performing menial service but not on the establishment of a unit, cf. Jones, 571, 585, 598;
in the army other ranks in waiting were adcrescentes. Suet. Claud. 25.1: instituit et
imaginariae militiae genus, quod vocabatur supra numerum, quo absentes et titulo tenus
Jungerentur, “he also instituted an imaginary type of military service, which was called
‘supranumerary’ and was performed titularly by absentees”, describing merely honorary
officer-status, offers essentially the same term in a sense apparently not exampled
elsewhere. Cf R.O. Fink, doc. 58, col. ii.l2, recording a single supranumerarf—]
belonging to a century of legio III Cyrenaica, c. AD 90, but his duties are obscure. At
111.20 V. uses the term to denote reserve forces on the battle-field, again in a non¬
technical way. The term was also used in the third century ad to distinguish centurions
assigned to special duties, cf J.R. Rea, “Ordinatus”, ZPE xxxviii (1980) 217-219.

yascicularia.

^muniftces.

^munera, cf 11.7 fin.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

20. Soldiers ought to deposit half their donative “with the
standards’% to be kept in savings for them.*

It was a divinely inspired institution of the ancients to deposit “with the
standards”^ half the donative^ which the soldiers received, and to save
it there for each soldier, so it could not be spent by the troops on
extravagance or the acquisition of vain things."* Most men, the poor
especially, spend as much as they can get.

This depositing of savings is judged to be in the interest first of the
troops themselves, for as they are maintained by the public
remuneration,^ half of each donative goes into increasing their personal
savings.^ Secondly, the soldier who knows that his spending money is
deposited “with the standards” never thinks of deserting, has greater
love for the standards, and fights for them more bravely in battle, since
it is human nature to care most about things on which one’s fortune is
staked.

So ten money-bags,^ that is, ten sacks, were laid down one for each
cohort, and in them was stored this account. An eleventh sack was also
added, into which the whole legion made a small contribution for burial
expenses. If any of the soldiers died, the cost of his burial might thus
be defrayed from this eleventh sack.* These accounts were preserved in

'On the regimental savings bank, cf. G.R. Watson, The Roman Soldier (London 1969)
104 ff., R.O. Fink, Roman Military Records on Papyrus (Cleveland, Ohio 1971) doc. 68,
esp. p. 245, M.P. Speidel, “The pay of the auxiHd‘\ Roman Army Studies I (Amsterdam
1984) 83-89, J.F. Gilliam, “The deposita of an auxiliary soldier", Mavors: Roman Army
Researches (Amsterdam 1986) 317-327, R.W. Davies, “A note on lorictitis'\ BJ clxviii
(1968) 161-165. See also R. Alston, “Roman military pay from Caesar to Diocletian”,
y/tyixxxiv (1994) 113-123.

^apud signa.

^donativum, bounty distributed on imperial birthdays and accession days, which by the
late-fourth century ad seems to have completely eclipsed the annual stipendium which was
by now of nugatory value, cf Jones, 623-624.

^Some soldiers got rich on the opportunities of army life, cf Amm. 22.4.7: ut per
ambitiones otiumque opibus partis auri et lapillorum varietates discerneret scientissime,
contra quam recens memoria tradidit (speaking of the corrupt court troops of Constantius
11), “so that having got rich as a result of corruption and idleness (the soldier)
distinguished the varieties of gold and jewels with expert skill,—a far cry from the recent
memory (of tough army life under Diocletian and Maximian).”

^annona, including food and drink, bedding and clothing, for which soldiers of the late
Empire, unlike their predecessors, did not have to pay.

^'castrense peculium, cf Dig. 49.17, 37.6.1.15, 22; Pauli Sent. 3.4a.3.

^folles.

”Cf Onas. 36.1-2 on the importance of proper burial arrangements.

BOOK II: THE ANCIENT LEGION

a wicker coffer, “with the standard-bearers”/ as they now say. For this
reason, standard-bearers were chosen not just for their trustworthiness,
but for being literate men too, who would know how to look after
deposits and render account to each man.

21. Promotions in the legion are made in such a way that those
promoted pass through all cohorts.

It was not by human counsel alone but by divine inspiration as well, in
my opinion, that the Romans organized the legions. In them the ten
cohorts were so arranged that they all seemed to be one body, one
unity.^ For soldiers are promoted as if in a circle through different
cohorts and different administrative departments,^ so that a man
promoted from the First cohort to any grade goes to the Tenth cohort,
and then comes back as his years of service increase with higher grades
through the other cohorts to the First again. Therefore it is in the First
cohort that the chief centurion"^ reaches this palm of honour in which he
of all the legion gains unlimited privileges, after he has administered all
cohorts in rotation through different departments, just as the First
secretary^ of the Praetorian Prefects’ staff reaches an honoured and
lucrative end to his service. In this way, through the bonds of the 10 -
man section,^ the legionary cavalry honour their own cohorts even
though horse are naturally inclined to be on bad terms with foot.^ By
means, therefore, of this interweaving in the legions of all cohorts and
of cavalry and infantry, one harmonious spirit was preserved.*

^apud signiferos. Apparently meaningless pedantry?

^Cf. Thuc. 2.11.9 on the Spartan military ideal put into the mouth of Archidamas.
'scholae.

^centurio primi pili. Cf, 11.8 in it. for the commoda primipilatus
^phmiscrinius.

^'contubernium. V.’s calculations do not allow for cavalry to be included among the
contubernia at 11.6 and 11.8

’On the modem antipathy between cavalry and foot, which probably arose because they
belonged to separate units, cf. Julian’s demotion of a cavalry regiment to infantry rank,
Amm. 24.5.10.

*The thought is that, unlike in the modem field armies, the infantryman promoted
cavalryman normally remained a member of the same legion, and might pass through a
succession of promotions within it. This was still true as late as the reign of Diocletian,
cf T. Drew-Bear, “Les voyages d’Aurelius Gaius, soldat de Diocl^tien”, in La Geographie
administrative et politique d'Alexandre a Mahomet, Actes de Colloque de Strasbourg
14-16 juin 1979 (Univ. de Strasbourg, 1982) 97, 101-102. More particularly, V. may
misunderstand a note that under the Principate legionary cavalry remained on the books
of the century of first enrolment, as appears from surviving evidence, cf D.J. Breeze,

VEGETIUS’ EPITOME OF MILITARY SCIENCE

22. The difference between trumpeters, hornblowers and the bugle
call.*

The legion also has trumpeters,^ hornblowers^ and buglers/ The
trumpeter calls soldiers to battle, and again sounds for a retreat. When
hornblowers sound, it is not the soldiers as such but the standards that
obey their signal. So when soldiers alone are due to go out to do some
work, the trumpeters sound. When the standards are to be moved, the
hornblowers sound. But when there is to be a battle, both trumpeters
and hornblowers sound together. The classicum is the name for the
signal sounded by the buglers on the “hom”.^ This is considered the sign
of the High Command,^ because the classicum is sounded when the
Emperor’ is present, or when capital punishment is being inflicted on a
soldier, since this must be done according to the Emperor’s laws.*

So when soldiers go out to perform night-watch duties and outpost
duties, to do some work, or drill in the parade-ground, they start work
at the call of the trumpeter, and stop again at the signal from the
trumpeter. But when the standards are moved or, once moved, are to be
planted, the hornblowers sound. The signals are observed in all exercises
and marches, so that soldiers may obey more readily in actual battle, if
ordered by the leaders to fight, stand their ground, pursue or retire. For

“The organization of the legion: the First cohort and the equites legionis'\ JRS lix (1969)
50-55. The ideal of promotion through all departments perhaps belongs to the civil
service.

'classicum.

^tubicines.

^cornicines.

^bucinatores.

^cornu = bucina, i.e. bugle, but V.’s logic here is obscure.

^'imperium.

^imperator, lit. Commander-in-chief, who was the Emperor.

"^Cf Caes. Bell. Civ. 3.82.1: Pompeius... suum cum Scipione hortorem partitur
classicumque apud eum carti et alterum illi iubet praetorium iendi. “Pompey divided his
command with Scipio and ordered the classicum to be sounded at his headquarters and
a second command-centre to be put up.” Ps.-Hyginus 21:5/ longiora fuerint fsc. castra),
classica dicentur nec bucinum in tumultu ad portam decimanam facile potuerit exaudiri.
“If the camp is too long, the classica will be sounded but in an uproar the bugle will not
be easily able to be heard at the rear gate.” The Emperor’s laws will be military
regulations, some of which may have been issued ad hoc as constitutiones. Cf A.
Neumann, “Das rOmische Heeresreglement”, CPh xli (1946) 217-225, = ditto, HZ clxvi
(1942) 554-562, but with addendum pp. 222-225, A. Neumann, “Das Augusteisch-
Hadrianische Armeereglement und Vegetius”, CPh xxxi (1936), 1-1. See also A.A.
Schiller, “Sententiae Hadrian! de re militari”, in Sein und Werden im Recht: Festgabe fur
Ulrich von Lubtowzum 70. Geburtstag am 21. August /97f? (Berlin 1970) 295-306, who
argues contra Neumann that there may well have been no rule-book as such.

BOOK II: THE ANCIENT LEGION

it is obviously a sound principle that they ought always to be doing in
peacetime what it is deemed necessary to do in battle.'

23. The training of soldiers.

Now that the organization of the legion has been summarized, we return
to the exercises from which, as has been said already,^ the army took its
name. Recruits and novice soldiers were trained morning and afternoon
in all types of arms, but veterans and trained soldiers also exercised with
their arms once a day without fail. For length of time or number of
years does not transmit the art of war, but continual exercise. No matter
how many years he has served, an unexercised soldier is forever a raw
recruit.

Armatura^ which is displayed on festal days in the Circus,used to
be learned not just by armaturae^ under the drill master,^ but by all
ordinary soldiers alike in daily practice. For speed is acquired through
bodily exercise itself, and also the skill to strike the enemy whilst
covering oneself, especially in close-quarters sword fighting. What is
more, they learn how to keep ranks and follow their ensign through
such complicated evolutions in the mock-battle itself No deviation
arises among trained men, however great the confusion of numbers.’

It is also very useful for them to exercise with the post and foils,
because they learn to go for the flank, feet or head with the point and
with the edge.® Let them grow used to executing jumps and blows at the
same time, rushing at the shield with a leap and crouching down again.

'Cf. III.5.

^Cf. II.1.

^special drill, cf, 1.13 note.

‘‘i.e. Circus Maximus in Rome. Cf. Claud, pan. de Hon. cons. VI 621-640 for
contemporary tournaments of the type celebrating the consular games of ad 404, and Livy
44.9.2-7 (169 BC) for the Republican tournaments in the Circus— ludicro circensi —put
on by iuvenes Romani. Also under the Principate, cf. Suet. d. Julius 39.3, Claud. 21.3,
Dom. 4.1, both pedestrian and equestrian tournaments, and the lusus Troiae whose origin
Vergil describes in Aen. 5.553-603.

^special light troops, trained in armatura\ cf. Firm. Mat, Math. 8.6.3: aut <qui in> dorso
Stans equorum mirifica se moderatione sustentet, atque adprime equo vectus militares
armaturas exerceat. “or <one who> balances himself with wonderful control standing on
the backs of horses, and riding his horse best of all trains the military armaturae."

^'campidoctor.

"Cf. 1.13 and 11.13.

^Despite 1.11, Polyb. 18.30.7 shows that the Roman ordinatio had soldiers using swords
for both cutting and thrusting, as here and at 1II.4.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

now eagerly darting forward with a bound, now giving ground, jumping
back.' Let them also practise hitting the same posts from a distance with
javelins, to increase their skill at aiming and the strength of the right
arm.^

Archers and slingers used to put up scopae, that is, bundles of
brushwood or straw, for a target, removing themselves 600 ft. (= c.580
ft., 177 m.) from the target, to practise hitting it frequently with arrows,
or stones aimed from a “sling-staff’.^ This enabled them to do without
nerves in battle what they had always done in exercises on the training-
field. They should also be accustomed to rotating the sling once only
about the head, when the stone is discharged from it. All soldiers also
used to practise throwing stones of 1 lb. weight by hand alone."* This
was considered a readier method, because it does not require a sling.^

They were also made to throw javelins and lead-weighted darts in
continual and perpetual exercises; so much so, that in winter-time they
built riding-schools^ for the cavalry and a kind of drill-half for the
infantry, roofed with tiles or shingles or, failing these, thatched with
reeds, sedge or straw. In them the army was trained in arms under
cover, when the weather was disturbed by wind and rain.* But for the
rest of the time, even in winter, so soon as snow and rain ceased, they
were made to train on the exercise-field, so that no interruption to
routine might weaken soldiers’ minds and bodies.

It is advisable that they should very frequently be felling trees,
carrying burdens,^ jumping ditches,'*^ swimming in the sea or rivers,"

'Cf. 1.9. Serv. ad Verg. Aen. 11.284: Quantus in clipeum adsurgat: aut "quantus" est
quotiens in hostem pergens erigit scutum: aut pugnandi exsecutus est genus. Qui enim
scripserunt de arte militari dicunt summum genus esse dimicandi, quotiens catcato umbone
adversarii se in hostilem clipeum erigit miles et ita contra stantis vulnerat terga. ‘“With
what force (Aeneas) charges the shield’: either quantus refers to when advancing against
the enemy he puts up his shield; or he executed a type of combat. For those who have
written on the art of war say that it is the best type of combat when a soldier steps on the
boss of his adversary and raises himself on the enemy shield and from this position
wounds the opponent’s back.”

^Cf 1.14.

^fustibalus, cf 111.14.

‘*1 Roman pound = 0-7219 lb. avp., or c. 1 VA oz. or 327-45 grammes, cf. 1.19 note.
^Cf. 1.15-16.

^'porticus

^basilicae.

“^Cf 1.18, III.2.

BOOK II; THE ANCIENT LEGION

marching at full step or even running in their arms, with their packs on.'
The habit of daily labour in peace may not then seem arduous in war.
Whether they be legion or aiailia, let them be training constantly. As
a well-drilled soldier looks forward to battle, so an untrained one fears
it. Finally, note that technical skill is more useful in battle than strength.
If training in arms ceases, there is no difference between a soldier and
a civilian.

24. Examples to encourage military exercises, drawn from other
arts.

Athletes, show-hunters and charioteers are accustomed to maintain and
improve their skills by daily practice for paltry profit or at least to gain
the favour of the populace. The soldier, by whose hands the State must
be preserved, ought to be keener in keeping up his knowledge of
fighting and practice of warfare by continual exercises. He wins not
merely a glorious victory, but even greater spoils, since the hierarchy of
the soldiers and the Emperor’s judgement regularly exalt him to riches
and dignities.^ Stage artists never stop rehearsing to win the praise of
the populace. The soldier, once he has been selected and sworn in,
ought not to stop exercising at arms whether a novice or even an old
hand. For he must fight for his own life and the liberty of all, and it is
above all the ancient and wise opinion that all arts depend on practice.

25. Catalogue of the legion^s tools and machines.

The legion is accustomed to be victorious not only on account of the
number of soldiers but the type of its tools also. Above all, it is
equipped with ballista-bolts^ which no cuirass or shield can withstand.
For each century customarily has its own carriage-ballista,'^ with mules
assigned to draw it and a section,that is, eleven men, to arm and aim

“'Cf. 1.9.

"Cf. 1.10.

'Cf. 1.19, 1.9, 1.27.

^Cf II.7 and 11.8.

^iacula.

^carroballista, cf. the ballista quadrirotis —four-wheeled balI ista—described and illustrated
by the Anon, de Rebus Bellicis 7, a design similar to those shown mounted on two¬
wheeled mule-carts or wooden pillar-bases on Trajan’s column, C'ichoriiis PI. XXXI, scene
xl, PI. XLVIl, scene Ixvi, and also to the miniature mode! in (Ps.)-Heron’s
Cheiroballistra, a late-Roman technical work. See further under IV.22

^contubernium.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

it.’ The larger these machines are the farther and more violently they
shoot projectiles. Not only do they defend a camp, but they are also
placed on the battlefield behind the line of the heavy infantry. Faced
with their attack, neither the cavalry cuirassiers nor the shield-bearing
infantry of the enemy can stand their ground. In one legion there are
traditionally 55 carriage-ballistas.^ There are also ten mangonels,^ one
for each cohort. They are carried around ready-armed on ox-carts, so
that should the enemy come to attack the rampart to storm it, the camp
can be defended with darts and rocks."*

The legion also carries with it canoes^ hollowed out of single logs,
and a supply of very long ropes and sometimes iron chains too. By
binding together these “single timbers”,^ as they are called, and throwing
planks on top, rivers that lack bridges and cannot be forded may be
crossed safely by both infantry and cavalry.^

The legion has grappling-irons* called “wolves”,^ and iron siege-
hooks’’^ fixed on the end of very long poles." Also, for constructing
earthworks, it has forks, mattocks, spades, shovels, troughs and baskets
for carrying earth. There are also axe-picks, axes, adzes and saws for
chopping and sawing timber and stakes.'^ It has craftsmen too, equipped
with every tool, making siege-sheds, mantelets, rams, “vines”," as they
call them, and mobile towers, for attacking enemy cities.'"* But to avoid
my speaking at too great length by listing each item separately, the
legion ought to carry with it everywhere all that is thought necessary to

'Cf. Maur. Strat. 12.B (8) 6, on the equipment (one wagon each) of each 10-man
contubernium. Significantly, some apparently had wagons with revolving ballistae at each
end. V.’s eleventh man is the “dean” or decanus/caput contubernii, cf. II.8.

^About the right number, according to Jos. Bel!. Jud. 3.166, where Vespasian’s three
legions brought to bear 160 pieces of artillery against Jotapata in ad 67.

^onagri, “wild asses”.

"Cf. IV.22.

^scaphae.

^"monoxyli, cf 111.7.

'Cf ni.7.

'^harpagones.

'^lupi.

'yalces.

"Cf IV.23.

•^Cf 1.24.

^\ineae, i.e. penthouses.

'"Cf 11.11, IV.14^17.

BOOK 11: THE ANCIENT LEGION

any kind of warfare, so that in whatever place it pitches camp, it makes
an armed city.'

‘Cf, II.18.

Book III

Preface.

The annals of old declare that the Athenians and the Spartans were
masters of the world before the Macedonians. But among the Athenians
throve the cultivation of other arts besides that of war, whereas the chief
concern of the Spartans was war.' Indeed, they are credited with having
been the first to collect evidence about the fortunes of battles and write
an art of war; insomuch, that they reduced warfare, previously thought
to be restricted to courage alone or at least luck, to a discipline and
study of skills, and they instructed drillmasters, whom they called
“tacticians”,^ to teach their youth the various fighting techniques.^ O
men worthy of the highest admiration and praise who wished to learn
that art in particular without which the other arts cannot be!"*

Following these men’s precedents the Romans maintained the
principles of warfare in practice and transmitted them in writing. This
material, dispersed through various authors and books. Invincible
Emperor, you ordered my Mediocrity to summarize,^ so that neither
should boredom arise from excessive detail, nor complete confidence be
lacking because of brevity. The extent to which military science was of
benefit in the battles of the Spartans is made clear from the case of
Xanthippus, not to mention the rest. When he brought help as an
individual to the Carthaginians not by courage but by skill, using armies
that had been utterly defeated, he captured and conquered Atilius
Regulus and an often victorious Roman army. By triumphing in a single

'On the Spartans being interested in nothing more than improving their skills at war, cf.
Plato Laches 182e-183b. On superiority of Spartan tactical training, Xen. Const. Laced.
11.510. On Spartan training, Arist. Pol. 8.4, 1338b.

^tactici

^Cf. E.L. Wheeler, “The hoplomachoi and Vegetius’ Spartan drillmasters”, Chiron xiii
(1983) 1-20. The identity of these tactici is unknown, unless the Arcadian (?) Aeneas
Tacticus, by far the most significant early tactical writer, is meant.

^Ael. Tact. 1.7 quotes Plato Leges 625e-626b for the same in praise of Cretan lawgivers,
there being by nature “undeclared war between all cities”. Cf also III. 10 init.

^Cf. Amm. 23.4.1: quantum mediocre potest ingenium, “as far as a modest mind is able.”

BOOK III: FIELD STRATEGY AND TACTICS

encounter, he concluded the entire campaign.' So also did Hannibal
obtain the services of a Spartan tactician, when he was going to invade
Italy. It was due to his advice that he destroyed so many consuls and
legions, though inferior himself in numbers and strength.^

Therefore, he who desires peace, let him prepare for war.^ He who
wants victory, let him train soldiers diligently. He who wishes a
successful outcome, let him fight with strategy, not at random. No one
dares challenge or harm one who he realizes will win if he fights."^

(Logistics, commissariat, discipline, signals,
castrametation, ch. 1-8)

1. The proper size of an army.

The First book set out the selection and training of recruits. The Second
explained the formation of the legion and its military disciplines. This
Third book sounds the classicum.^ For the former matters were
discussed first so that the present subject, which comprises the skills of
engagement and the elements of victory, may preserve the order of the
discipline, thus being more readily comprehensible and of greater
assistance.

'A Spartan general called in as adviser by the Carthaginians after their defeat on African
soil by M. Atilius Regulus, cos., in 256 BC, Xanthippus reformed the Carthaginian army
and defeated and captured Regulus the following year, whereupon he left Carthaginian
service; but the war was eventually won by Rome in 241 BC. Cf. App. Lib. 3, Cass. Dio
fr. 43, 24, Polyb. 1.32 ff, Diod. Sic. 23.14 ff., Cic. de Off. 3.99, Florus 1.18.23: nam
conversis ad extrema auxilia hostibus, cum Xanthippum illis ducem Lacedaemon mississet,
a viro militiae peritissimo vincimur “The enemy turned to desperate remedies, Sparta sent
them Xanthippus as a general, and we were defeated by a man most expert in warfare.'’

^Identified by Wheeler, art. cit. 1, as Sosylus. The surviving fragment of his history of
Hannibal shows deep historical knowledge of tactics, cf U. Wilcken, “Ein Sosylus
Fragment in der Wurzburger Papyrussammlung”, Hermes xW (1906) 141. F. Jacoby, FGH
IIA(1930) no. 176, id., IIIA.l (1929) 1204, seems to underestimate his status. Nepos
Hann. 13.3 says that he taught Hannibal Greek.

^The most memorable phrase in all Vegetius, and the earliest expression in Latin of the
classic paradox, si vis pacem, para be Hum. “If you want peace, prepare war.” W. Haase,
“Si vis pacem, para bellum”, in Akten des XI. internationalen Limeskongresses 1976
(Budapest 1977) 721-755, explains it as deterrence-theory, but the Thuc. 4.92,5 parallel
suggests that it need not exclude the pre-emptive strike. Cf also A. Otto, Die Sprichwdrter
undsprichwdrtlichen Redensarten der Romer (Leipsig 1890) 54 s.v. “bellum”, with refs,

'Cf IV.31. Cf also Jos. Bell. lud. 3.72.

^bugle-call of the High Command, cf 11.22.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

“Army”' is the name for a host of legions, auxilia and cavalry
gathered together for the purpose of waging war.“ Its proper size is
discussed by military experts. For when one reads the examples of
Xerxes,^ Darius,"* Mithridates^ and other kings who armed countless
populations, it is clearly apparent that over-large armies have been
overcome more by their own size than the bravery of the enemy. For a
greater multitude is subject to more mishaps. On marches it is always
slower because of its size; a longer column often suffers ambush even
by small numbers; in broken country and at river-crossings it is often
caught in a trap as a result of delays caused by the baggage-train. Also
it is an enormous labour to collect fodder for large numbers of animals
and horses. Difficulties again with the grain-supply, to be avoided on
any expedition, afflict larger armies sooner.^ For however thoroughly
rations may have been prepared, they run out more quickly, the more
they are distributed to. Finally water itself sometimes hardly suffices for
too large a number.^ And if for some reason the battle-line should turn
tail, more casualties must inevitably occur to more men, and those who
escape, once terrified, thereafter fear battle.

'exercitus.

^The tripartite field army of the fourth century ad is meant, cf. Hoffmann, 1.72, 397.

'Xerxes, king of Persia 486-465 BC, invaded Greece 480 with allegedly 1,700,000
infantry, 1,207 warships, 3,000 smaller ships and 80,000 cavalry, and obtained large
additional forces en route from the Hellespont, cf Hdt. 7.60 ff, 184 ff., but what modem
historians estimate as totalling nearer 180,000 men in his army and 800 triremes, cf J.B.
Bury and R. Meiggs, A History of Greece (London 1975'^) 169. They were defeated at sea
at the battle of Salamis 480 and on land at the battle of Plataea 479 BC.

‘‘Darius I, king of Persia 521-486 BC, father of Xerxes, sent an expedition to Greece 490,
defeated on land at the battle of Marathon 490.

'Mithridates VI Eupator, king of Pontus, assembled a huge force against Lucullus in 73
BC but was defeated at Cyzicus by problems of supplies, exploited by his opponent; cf
App. Mithr. 69-78, Plut. Luculi 7-13.

^’Cf Pac. Pan. Lat. 2 (12) 32.5, on Theodosius I’s expedition against Maximus ad 388,
notable for the unprecedented size of its barbarian contingents. According to Pacatus, they
cheerfully shared out inadequate rations.

^Famously, Xerxes’ army drank numerous rivers dry, cf Hdt. 7.21.1, 43.1, 58.3, 108.2,
127.2, 196. Herodotus was less surprised that rivers ran dry than that grain-suppplies held
out, 7.187.

BOOK III: FIELD STRATEGY AND TACTICS

But the ancients, who learned to remedy their difficulties from
experience, wished to have armies that were not so much numerous as
trained in arms. So for smaller wars they thought one legion with mixed
auxilia could suffice, that is, 10,000 infantry and 2,000 cavalry; and this
force was often led on campaign by praetors, like lesser generals.' But
if the enemy’s numbers were said to be large, a man of consular
authority was sent with 20,000 infantry and 4,000 cavalry, like a greater
“counf’.^ But if a countless horde of the fiercest tribes had rebelled,
then under press of extreme urgency two generals and two armies were
sent, with the following instruction: “Let both consuls, jointly or
severally, provide that the Republic take no harm”.^ For when the
Roman People were fighting virtually every year in various regions
against different enemies, supplies of troops were adequate only because
they judged it more useful to have not so much large armies as more of
them,—yet the principle was observed that there should never be a
greater number of Allied auxiliaries in camp than Roman citizens."*

2. How the army’s health is controlled.

Next 1 shall explain a subject to which special thought must be
devoted—how the army’s health is preserved; that is, by means of site,
water-supply, season, medicine and exercise. By “site” I mean that
soldiers should not camp in pestilential areas near unhealthy marshes.

\elut minores duces: duces often means limitanean area commanders (“dukes”), but here
it is used non-technically, note velut, to denote lesser field-army generals— comites rei
militaris. The latter could not have been called minores comites by V. since this
technically denoted “dukes”, who ranked as Counts of the Second class, cf. T. Mommsen,
“Das rOmische Militarwesen seit Diocletian”, Hermes w'w (1889) 267 = id., Gesammelte
Schriften VI (Berlin 1910) 272.

modern Field-Marshal or magisier militum is meant, cf. also 11.9. Cf. Polyb.
3.107.10-15 for the old consular armies of the Republic.

The senatus consultum de republica defendenda or SC "uliimum ”, allegedly used in 464
BC (Livy 3.4.9) and 384 BC (Livy 6.19.3). However, it cannot be traced reliably before
133 BC against Ti. Gracchus (Val. Max. 3.2.17, Plut. Ti Gracch. 19), cf RE Suppl. VI
(1935) 756, and many would see its first certain use only in 121 BC by the consul L.
Opimius. V. gets the formula right, cf Cic. Phil. 8.14, Sail. Cat. 29.2, Caes. Bell. Civ.
1.5.3, but wrongly retrojects it into the context of Rome’s foreign, rather than civil, wars.
It would have been more correct to speak of the Republican dictatorship at this point. V.,
or a pro-senatorial source such as Frontinus, may have substituted the consular emergency
powers.

^sociales auxiliares. Cf Polyb. 6.26.7-8, 30.2.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

nor in arid plains and hills, lacking tree-cover, nor without tents in
summer. They should not move out too late in the day and fall sick
from sunstroke and marching-fatigue, but rather start a march before
dawn, reaching the destination in the heat of the day. They should not
in severe winter weather march by night through snow and ice, or suffer
from shortage of firewood or an inadequate supply of clothes. For a
soldier who is forced to be cold is not likely to be healthy or fit for an
expedition. Neither should the army use bad or marsh water, for bad
drinking-water, like poison, causes disease in the drinkers. Besides, it
requires constant vigilance on the part of officers' and tribunes and of
the “count” who holds the senior command to see that ordinary soldiers
who fall sick from this cause may be nursed back to health with suitable
food^ and tended by the doctors’ art. It is hard for those who are
fighting both a war and disease.

But military experts considered that daily exercises in arms were
more conducive to soldiers’ health than doctors. So they wished that the
infantry be trained without cease, under cover when rainy or snowing,
in the exercise-field on the rest of the days.^ Similarly they gave orders
that the cavalry should constantly train themselves and their horses not
only on the flat, but also over precipitous places and on very difficult
ways with gaping ditches, so that nothing unfamiliar might meet them
in the stress of battle.'* From this it is appreciated how zealously an
army should always be trained in the art of war, since the habit of work
may bring it both health in the camp, and victory in the field.

If a multitude of soldiers stays too long in autumn or summer in the
same place, then drinking-water contaminated by pollution of the water-
supply and air tainted by the foul smell itself give rise to a most deadly
disease.^ This can only be prevented by frequent changes of camp.

^principia.

^e.g. chicken, cf. IV.7.

'Cf. 11.23.

'‘Cf. 1.27. V. ought perhaps to have mentioned the indoor riding-schools again, as at 11,23,
cf. 1.18, but his eye seems to have slipped over his source.

^i.e. cholera or typhoid. On the foul smell cf Sail. Jug. 44.4, Onas. 9.1.

BOOK III: FIELD STRATEGY AND TACTICS

3. How much attention should be devoted to the procurement and
storage of fodder and grain.

The order of subjects demands that I speak next about the provisioning-
system for fodder and grain.' For armies are more often destroyed by
starvation than battle, and hunger is more savage than the sword.^
Secondly, other misfortunes can in time be alleviated: fodder and grain
supply have no remedy in a crisis except storage in advance.^

On any expedition the single most effective weapon is that food
should be sufficient for you while dearth should break the enemy.
Therefore, before war is commenced, careful consideration should be
given to supplies and their issue in order that fodder, grain and the other
army provisions'' customarily requisitioned from provincials may be
exacted in good time, and quantities always more than sufficient be
assembled at points well-placed for waging war and very well-fortified.^
But if the taxes in kind be insufficient, everything (needed) should be
compulsorily purchased from advance payments in gold. For there is no
secure possession of wealth, unless it be maintained by defence of
arms.^

Often an emergency is doubled and a siege becomes longer than
expected,^ when the opposition though hungry themselves do not give
up besieging those whom they expect to be overcome by hunger. Also

'Cf Jones, 626-629.

^Front. Strat. 4.7.1, cf. Caes. Bell Civ. 1.72.1, Amm. 25.7.4.

'Cf IV.30.

^annonariae species.

^e.g. the huge d^pot at Batnae, where 50 men were killed by a collapsing pile of bales of
chaff, cf Amm. 23.2.8. Provincials had to contribute a wide variety of foodstuffs and
commodities as taxes in kind for delivery to government warehouses specifically to supply
the army, in the system known as annona militaris, cf Jones, 458—460, 626-630, etc.

^’Cf Amm. 17,3.1 ff. for conquisita or incrementa —special supplementary levies an¬
nounced by the Praetorian Prefect Florentius, to make up what was lacking from the poll-
tax and land-tax accounts. V. seems to envisage mandatory pre-payments of tax.

^V. unconsciously assumes that warfare will be defensive, that campaigns will be on
Roman territory. This happened, for example, in ad 377 when Adrianople was besieged
by Fritigern’s Goths and the rich farmlands of Thrace overrun, cf Amm. 31.6. The
Romans responded by driving the marauders into the mountains, where they
unsuccessfully attempted to starve them into submission or death, cf Amm. 31.8.1, as
they had the Isaurians in AD 354 and 367-368, cf Amm. 14.2.13, 27.9.7.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

all livestock, any sort of fruit and wine which the enemy invader can
seize for his own sustenance should be collected into strong forts
secured by armed garrisons, or into very safe cities, by landowners
acting under the admonition of edicts or the compulsion of specially
appointed escorts,' and the provincials impelled to shut themselves and
their property behind fortifications before the invasion." Repairs to all
walls and torsion-engines should be taken in hand in advance too, for
if the enemy once find you unready, everything becomes confused in
panic and things needed from other cities are denied you through the
roads being closed. Faithful stewardship of granaries and controlled
issue usually provides for a sufficiency, especially if taken in hand from
the outset. But economy comes too late to save (grain) when there is a
deficiency.^

On arduous campaigns the ancients used to provide rations by heads
of soldiers rather than by status, on the understanding that after the
emergency there was restitution to these men" from the State.^ In winter
problems of firewood and fodder, in summer of water should be
avoided. Shortages of grain, wine-vinegar, wine and salt should be
prevented at all times. Therefore cities and forts should be defended by
those soldiers who prove less useful in the field, equipped with arms,
arrows, “sling-staves”,^ slings and stones, mangonels^ and catapults.®

^prosecutores.

^Cf. measures taken in ad 359 in response to king Sapor’s invasion of Mesopotamia: the
compulsory movement of the population, rich and poor, into fortified strong-points, along
with the stockpiling therein of supplies and collection of livestock, and the burning of the
country abandoned; Amm. 18.7.3. Also cf. similar strategy against the Goths overunning
Thrace in ad 376-377, Amm. 31.81.1. The historic precedent was set by Q. Fabius
Maximus (Cunctator), dictator in 217 BC, cf Livy 22.11.4.

'Cf 1V.7.

^i.e., senior ranks.

^This step was taken by Julian’s army in the middle of the Persian campaign in AD 363,
cf Amm. 25.2.1.

^'fustibali.

^onagri^ “wild asses”.

*ballistae. V. may possibly refer to limitanei or frontier garrison troops, and burgarii or
civil guard. The system of fortresses and staging-posts, themselves often doubling as
cities, through which the expeditionary army would move until it reached enemy territory
is described by SKA Alex. Sev. 45.1-3, 4T1 and Ambr. Exp. Psalmi CXV/If 5.2. The
stingstaves are discussed at III.14, the mangonels and catapults at IV.22.

BOOK III: FIELD STRATEGY AND TACTICS

Especial care should be taken lest provincials in their unsuspecting
simplicity be deceived by the treachery and perjuries of the enemy.
Pretended trade and peaceful relations have more often caused harm to
gullible people than arms. By this strategy the enemy if they collect
together suffer famine, and if they disperse are easily beaten by frequent
surprise attacks.’

4. Measures needed to ensure that soldiers do not mutiny.^

An army gathered together from different places occasionally raises a
riot and, when in fact it is unwilling to fight, it pretends to be angry at
not being led out to battle. This is chiefly the action of those who have
lived at their home base in idleness and luxury. Taking offence at the
harshness of the unaccustomed effort which it is necessary to endure on
campaign, fearing battle besides, having shirked exercises in arms, they
plunge headlong into a rash enterprise of this sort.

A compound treatment is usually applied to this wound. While they
are still separate and in their base, (soldiers) should be held to every
article of discipline by the strictest severity of tribunes, “vicars”^ and
officers,"' and observe nothing but loyalty and obedience. They should
be doing campicursio,^ as they themselves term a review of arms,
constantly, they should have opportunity for no leave of absence, they
should continually be obeying the muster*" and be present at the

'The strategy which Fritigern had to face in AD 377, Amm. 31.11.5, cf. 31.7.6. It was
also that endorsed by Ammianus himself, 31.7.2.

^The chapter has some resonances with Polyb. 11.25-30, in which Scipio Africanus Maior
quells a mutiny, cf Livy 28.24-29. V. takes the opportunity to repeat again a summary
of the training, cf 11.23.

\ tear a, acting tribunes.

*principia.

^parade-ground marching-manoeuvres. Cf III.9 fm., where it seems to mean “field
manoeuvres’' as for a battle, and 11.22 decursio campi, “parade-ground drill”.

^'ad nomen observare, cf ad nomen respondere in Livy 7.4.2, 2.28.6, 28.29.12; cf also
Maur. Strat. 7 (A) 2, dSvoupid^Eiv, v.l. for dyidi^eiv (td pdvSa). The Byzantine
dSvodpiov denoted a meeting of the troops for a blessing, address, counting and
distribution of pay, cf H, Mihaescu, “Les 6I^mcnts latins des Tactica-Strategica de
Maurice-Urbicius et leur 6cho en Neo-grec”, RESE vi (1968) 497. It could also be
dangerous to the High Command to assemble the troops en masse so as to realize their
strength, cf Leo Tact. 9.4, 6.15. But there were also smaller assemblies for organizing
training and manoeuvres, etc., such as V. must mean, cf Du Cange (ed.), Glossarium ad

VEGETIUS’ EPITOME OF MILITARY SCIENCE

standards,' and be kept as frequently as possible shooting arrows,^
throwing javelins,^ throwing stones with the sling or by hand/
performing the gestures of the armatura,^ fencing with foils made to
imitate swords with the point and with the edge for most of the day
until they are exhausted/’ They should furthermore be trained at leaping
over fosses by running and jumping/ If the sea or a river is near their
base, in summer they should all be made to swim/ also to fell trees,^
march through thickets and broken country,'® hew timber, open a fosse,"
occupy some point, and strive with shields mutually opposed not to be
dislodged by their comrades.'^ Soldiers who have been so trained and
exercised at their base, whether they are legionaries, auxilia or cavalry,"
when they come together for a campaign from their various units
inevitably prefer warfare to leisure in the rivalry for valour. No one
thinks of mutiny, when he carries confidence in his skill and strength.

But the general should be careful to learn from tribunes, “vicars” and
officers in all legions, auxilia and vexillations,"*—not according to the
malice of informers but the true facts—if there are any disorderly or
mutinous soldiers. The more prudent policy is then to segregate them
from camp to so some work which might seem to them almost desirable
or else to allocate them to fortifying and guarding forts and cities, with
such subtlety that they seem to have been specially selected although
they are being cast off. For an army never breaks out in dissent with

scriplores mediae el infimae Graecilatis (Lyon 1688) s.v,

'Cf. Maur. and the ceremony alluded to above.

^Cf, 1,15.

’Cf. 1,14.

’Cf 1.16.

’special drill, cf 1.4. 1.13.

‘reading ad vectes... feriendos. Cf 1.11-12.

’Cf 1,9.

*Cf l.IO.

■’Cf 11.23.

"’Cf 1,9, 1.27.

"Cf 1.21-25, III.8.

"Cf 1.9,

"The three divisions of the fourth-century field army; cf Hoffmann, 1.72, 397.
"“Vexillations” refers to the cavalry arm of the tripartite field army.

BOOK III: FIELD STRATEGY AND TACTICS

equal enthusiasm, but is incited by a few who hope to escape
punishment for vices and crimes by involving large numbers in
wrongdoing. But if extreme necessity urges the medicine of the sword,
it is juster to follow ancestral custom and punish the ringleaders of
crimes, so that fear extends to all, but punishment to few. However,
those generals who have instilled discipline in their army through hard
work and routine are more praiseworthy than those whose soldiers are
forced into submission by fear of punishment.*

5. How many kinds of military signals there are.^

Many indeed are the orders to be given and obeyed in battle, since no
remission is granted to negligence when men are fighting for their
lives.^ But of all the rest there is nothing so conducive to victory as
heeding the warnings of signals. Since an army in the confusion of
battle cannot be governed by a single voice, and many orders have to
be given and carried out on the spur of the moment in view of the
urgency of events, ancient practice of all nations devised a means
whereby the whole army might recognize by signals and follow up what
the general alone had judged useful.

So there are generally agreed to be three types of signals, voiced,
semi-voiced and mute. Of these the voiced and semi-voiced are
perceived by ear, whereas the mute are transmitted to the eye. Those
called “voiced” are pronounced by the human voice, such as a
watchword on night-watch duties or in battle, for example, “victory”,"*
“palm”,^ “virtue”,^ “God with us”,^ “Triumph of the Emperor”® and
whatever others the supreme commander in the army may choose to

'Sixth-century examples of such praiseworthy generals include Justinian son of Germanus
(Theophyl, 3.12.7), Maurice (Men. Prot. fr. 23.2-3, Blockley), and Philippicus (Evagr.
6.3, Theophyl. 2.4.3-4).

traditional chapter in the tactici, cf. Arr. Tact. 27, Ael. Tact. 35.

'Cf. 1.13 fin.

^victoria.

^palma.

^Wirtus.

^Deus nobiscum.

*triumphus imperatoris.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

give. But note that these words should be changed daily, lest the enemy
recognize the sign from familiarity and spies pass among our men with
impunity.'

The “semi-voiced” are those given by the trumpet,“ hom^ or bugle.'^
The “trumpet” is the name for the straight instrument. The “bugle” is
that which is bent back on itself in a bronze circle. The “horn” is that
which is made from the wild aurochs, bound with silver, and when
modulated with a skilful breath emits a note of singing wind.^ By these
means through unambiguous sounds the army recognizes whether it
should halt, advance or retreat, whether to pursue fugitives into the
distance or sound for a withdrawal.

The “mute” signals are eagles, dragons, tx\s\gx\s,flammulae, tufae and
plumes.^ The soldiers accompanying their standard must go wherever
the general directs them to be carried. There are also other mute signals
which the general gives orders to be kept on horses, on clothes, or on
the arms themselves, to distinguish them from the enemy. Besides this
he may indicate something with his hand or, in barbarian fashion, with
a whip, or even by a movement of the clothes he is wearing.’ All this,
in camp, on the march, in every field exercise, every soldier should
learn to follow and understand. For continual practice is obviously

‘Cf. Onas. 26, and III. 26.

^tuba.

^cornu.

^bucina.

^Cf. Hieron. in Os. 5.8: bucina pastoralis est et cornu recurvo efficitur. “The bugle is a
shepherd’s instrument made from a curving horn.” R. Meucci, “Lo strumento del
bucinator. A. Surus e il cod. Pal. Lat. 909 di Vegezio”, BJ clxxxvii (1987) 259-272
proposes that in V. the words bucina and cornu have become transposed; cf. 11.22. See
also P. Barton, “On making a Roman cornu”, in Roman Military Equipment. The
Accoutrements of War, Procs. of the Third Roman Military Equipment Research Seminar,
ed. M. Dawson, BAR Int. Ser. 336 (Oxford 1987) 28-37.

^’Cf II.l, and loh. Lydus de Mag. 1.8: 56paTa ^TiipT^Kri... dKpo^i(|)i'5aq oOk
2X0via, ijcoprip^vaq X.o<|)idq* KaXov5m bt aOtdq oi ‘Pcopafoi ioOpaq, oi bt
pdppapoi To\3(|)a(;, ppaxd x\ mpa(|>0apei'ari(; tqf; pn4^^^ot Ttpoq toOtok;

oiovei 56paTa paxpd ^^riptrip^vcov v!)<l)aapdTcov—(j)XdppouX.a adtd dTcd toO
cl)XoYi vou xpco paxoq KaXouai, “long lances... having not bladed tips but crests rising in
the air; the Romans call them iubae, the barbarians tufae, corrupting the word a tittle; and
vexilla (ensigns) in addition to these, that is, long lances with cloths hanging from
them—(the Romans) call them flammulae from their flame colour.”

^Amm. 18.6.13 (movement of cloak), Claud, pan. de Hon. cons. VI 625 (whipcrack).

BOOK III: FIELD STRATEGY AND TACTICS

necessary in peacetime of a procedure which is to be maintained in the
confusion of battle.

There is also a “mute” signal common to both sides, when dust is
disturbed by an army as it marches, and rises up in clouds, betraying the
approach of the enemy. Similarly when forces are divided, they use fires
by night and smoke by day to signal to their comrades what cannot be
announced by other means.’ Some hang beams on towers of forts and
cities, indicating what is going on by now raising, now lowering them.^

6. The degree of caution to be observed when an army moves in the
vicinity of the enemy.

Those who have made a careful study of the art of war^ assert that more
dangers tend to arise on the march than in battle itself. For in battle
everyone is armed, and they see the enemy at close quarters and come
mentally prepared for fighting. On the march, the soldier is less armed
and less alert; he is thrown into instant confusion by a sudden attack or
concealed ambush. Therefore the general should take steps with ail
caution and prudence to ensure that the army suffer no attack on the
march, or may easily repel a raid without loss.

First, he should have itineraries of all regions in which war is being
waged written out in the fullest detail, so that he may learn the distances
between places in terms of the number of miles and the quality of roads,
and examine short-cuts, by-ways, mountains and rivers, accurately
described. Indeed, the more conscientious generals reportedly had
itineraries of the provinces in which the emergency occurred not just
annotated but illustrated as well, so that they could choose their route
when setting out by the visual aspect as well as by mental calculation,"*

'Onas. 6.8, cf. Hanno’s smoke-signal to Hannibal, Livy 21.27.7.

^Only very simple signals are meant, not a semaphore system. Cf. Onas. 25.3 on fire-
signals.

^V. never claims expertise; cf 1.8, 11.3.

^Note that the use of coloured maps was not a novelty in the time of V. Cf O.A.W.
Dilke, Greek and Roman A/aps (London 1985) 112, 210, R. Sherk, “Roman geographical
exploration and Military maps”, ANRW II. 1 (1974) 558 ft Cf Polyb. 9.13.6 on the second
principle of generalship—accurate geographical knowledge.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

In addition, he ought to find out everything from intelligent men,
from men of rank,' and those who know the localities, individually, and
put together the truth from a number of witnesses. Furthermore he
should collect^ at the risk of those responsible for choosing them^ able
guides, knowledgeable of the roads, and keep them under guard having
given them a demonstration of punishment and reward. They will be
useful when they understand that there is no longer any chance of
escape for them, and that there is ready reward for loyalty and
retribution for treachery. He should also make sure that men of
discernment and experience are found, lest the error of two or three
individuals put everyone at risk. Occasionally inexperienced rustics
promise more than they can deliver and believe they know what in fact
they do not."*

But the most important thing to be careful about is to preserve
secrecy concerning the places and routes by which the army is to
proceed. The safest policy on expeditions is deemed to be keeping
people ignorant of what one is going to do. It is for this reason that the
ancients had the standard of the Minotaur in the legions. Just as he is
said to have been hidden away in the innermost and most secret
labyrinth, so the general’s plan should always be kept secret.^ A safe
march is that which the enemy least expect to be made.

Nevertheless, some words should be said about how one ought to go
about meeting an attack, because scouts^ sent from the other side can

'honorati
^reading percipere.

heading eligentium: V. may have in mind cases of collusion between false guides,
working for the enemy, and barbarian Roman officers, such as was suspected of Latinus,
comes domesticorum, Agilo, tribunus stabuli, and Scudilo, tribune of the Scutarii, with an
unnamed Alaman guide in AD 354, cf. Amm. 14.10.8. Cf. Onas. 10.15, Polyb. 9.14.1-3.

‘’V. is consistently contemptuous of the intelligence of common people; cf. 111.3 fin.,
IV.41 fm., 1.3 fin.

^Paul. Fest. 135.21 L.: Minotauri effigies inter signa militaria esf, quod non minus occulta
esse debent cons ilia ducum, quam fuit domic ilium eius labyrinthus. “The image of the
Minotaur is included among the military standards, because the plans of the generals
should be not less secret than was the labyrinth, his home." Cf Onas. 10.22-24, Polyb.
9.13.2-5 on the first principle of generalship—secrecy, keeping one’s plans to oneself. On
the Minotaur and other early Roman standards, cf. Marquardt, 354 n. 4.

^'exploratores.

BOOK III: FIELD STRATEGY AND TACTICS

detect an expedition by its tracks or by sighting it, and occasionally
deserters and traitors are not wanting. When a general intends to set out
with his army in column, he should send ahead very reliable and quick¬
witted men on excellent mounts to reconnoitre those places through
which the army is due to march, both in advance and in the rear, and
to right and left, to prevent the enemy laying ambushes. Scouts operate
more safely at night than in daytime. In some measure a general betrays
himself if his scout* is captured by the enemy."

So let the cavalry take the road in front, then the infantry, with the
baggage,^ pack-horses,"* servants^ and vehicles placed in the middle, and
the light-armed portion of the infantry and cavalry bringing up the rear.
For attacks on a marching army are sometimes made at the front, but
more usually in the rear. The baggage-train should also be enclosed on
the flanks with equal strengths of soldiers, for ambushers frequently
attack the sides. But the part which the enemy is expected to approach
one should be particularly careful to reinforce with a screen of picked
cavalry, light-armed infantry and foot-archers. If the enemy surround on
all sides, reinforcements must be prepared on all sides. To prevent
added losses from a sudden commotion, soldiers should be warned
beforehand to be mentally prepared and have their arms in their hands.
In an emergency sudden things are terrifying, things that are foreseen
do not usually strike panic.^

The ancients took very thorough precautions against disturbance to
the fighting troops by servants getting wounded on occasion or afraid
or by pack-animals terrified at the din of battle, lest being extended too
far or massed together more than expedient, (the troops) might impede
their own side and help the enemy. Therefore they decided to marshall
the baggage-train like the soldiers under certain standards. So they

^speculator

^Cf. Maur. Strat. 7. B 13, 9.5.

^impedimenta.

^sagmarii.

^calones

^‘Cf. Onas. 6.5-6, Arr. Ect. 1-10, Jos. Bell. lud 3.115-126.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

selected men of ability and practical experience from among the
servants, whom they call galearii,^ and put them in charge of up to 200
pack-animals and grooms.^ To them also they gave ensigns, so that they
might know to which standards they should gather the baggage. But the
fighting men were divided from the baggage-train by a certain interval,
so that they were not pushed together and wounded in battle.

When an army is marching, the system of defence varies with
changes in terrain. On open plains cavalry are more likely to attack than
infantry, while on the contrary in wooded, mountainous or marshy
country, infantry forces are more to be feared. One thing to avoid is the
column being severed or thinned out through the negligence of one
group setting a fast pace while another is moving more slowly, for the
enemy immediately penetrate any gaps. Therefore the most experienced
drill masters,^ “vicars”"* and tribunes should be put in charge, with orders
to slow down those who are too brisk and force those going too slowly
to speed up. When an attack happens, those who have gone far ahead
wish to get away rather than go back. Meanwhile those who are in the
rear deserted by their comrades are overwhelmed by the violence of the
enemy and their own despair.^

One should bear in mind that the enemy sets up concealed ambushes
or engages in open battle only in places he thinks favourable to himself.
The general’s diligence provides against suffering damage from
surprises, so it is advised that he reconnoitre everything in advance.
Then if an ambush is detected and properly surrounded, it suffers more
damage than it was preparing to inflict.^ Again if an open battle is being
prepzired in mountain-country, the higher ground should be seized by

‘Cf. MO: lixas, quos gaiearios vacant. V. seems to mean a generic term for soldiers’
servants, rather than specifically those put in charge of the pack-animals and grooms. Cf.
M.P. Speidel, “The Soldiers’ Servants”, Anc.Soc. xx (1989) 239-248, = Roman Army
Studies If, 342-352.

^pueri, “boys”.

^campidoctores
\icarii, acting-tribunes.

^V. updates the centurions of the source to campidoctores, etc., as at 111.8. Cf. Jos. Bell,
lud 3.124, Arr. Ect. 9-10.

'’Cf. Onas. 6.7.

BOOK III: FIELD STRATEGY AND TACTICS

sending forces ahead so that when the enemy arrives, he finds himself
on lower ground, and dare not attack when he can see armed men in
front of him and overhead. But if there are routes which are narrow but
safe, it is better for soldiers to go ahead with axes and picks, opening
a road with their toil, than to suffer peril on the best route.'

We ought also to know the habits of the enemy—whether they
usually attack by night, at daybreak or during the rest-hour when men
are tired—and avoid that which we think they will do from routine.^ It
is likewise in our interest to know whether they are stronger in infantry
or cavalry, in pikemen or archers, and whether they are superior in
numbers of men or military equipment, so that we may adopt the tactics
which are judged^ useful to ourselves and disadvantageous to them.'' We
should calculate whether it is advantageous to begin the march during
the day or by night and how great are the intervals between places to
which we wish to advance, so as to save the men on the march from
being troubled by lack of water in summer, or faced with difficult or
impassable morasses and great torrents in winter, or the army from
being cut off before it can reach its destination through its march being
impeded.^

Just as it is to our advantage to avoid these things by being prudent,
so we ought not to let slip any opportunity which the enemy’s
inexperience or negligence offers to us. We should reconnoitre
assiduously, sollicit traitors and deserters so we can find out the
enemy’s present and future plans and, with our cavalry and light
armament in prepared positions, catch them in unforeseen ambushes
while marching or seeking fodder and food.

■Cf Jos. Bell lud. 3.118.

^Cf. III.9, IV.27.

^Docetur in this sense is a late-Latinism, also common in legal or otTicial documents, cf.
TLL s.v. doceo, 1708.59 ff., 1713.30 ff.

'Cf. III.9.

'Cf. m.2.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

7. How to cross large rivers.'

When crossing rivers careless armies often get into serious difficulties.
For if the current is too strong or the river-bed too wide it is likely to
drown baggage-animals, grooms and sometimes even the weaker
warriors. So when a ford has been reconnoitred two lines of horsemen
on picked mounts are lined up in parallel with sufficient space between
them for infantry and baggage-train to pass through the middle. The
upper line breaks the force of the waters, while the lower line collects
up any who may be snatched away or swept under, and brings them
safely across.^ But where the water is too deep to allow either infantry
or cavalry to cross, if the river flows through flat country, it may be
dispersed by digging multiple channels and easily crossed when divided.
Navigable rivers, however, are made passable by driving in piles and
boarding over the top, or else, for a temporary work, empty barrels may
be tied together and timbers placed upon them to provide a passage.^
Also the cavalry are accustomed to take off their accoutrements and
make fascines from dry reeds and sedge and place upon them cuirasses
and arms, so as not to get them wet. They and their horses swim across,
drawing <on reins>'' the fascines that they have tied to themselves.

But it has been found better for an army to carry around with it on
carts “single timbers”,^ which are rather wide canoes, hollowed out of
single trunks, very light because of the type and thinness of the wood.
Planks and iron nails are also kept with them in readiness. The bridge

'Cf. RE XX\ (1952) 2437-2452 s.v. “Pons” (Lammert), Maur. Strat. 1.9.8, 8.1.19, 9.1,11,
^The method was Caesar’s, cf. id. Bell. Civ. 1.64.5-6.

^The latter, presumably, for a kind of pontoon-bridge. V. does not mention the use of
inflated skins to buoy timber raft-bridges, cf J. Hornell, “Floats and buoyed rafts in
military operations”, Antiquity xix (1945) 73-79, or the bridge of allegedly skins and
cables alone described by the Anon, de Rebus Bellicis 16, carried on fifty packhorses or
the backs of “very few men” {praef. 14). Amm. 24.3.11 mentions skins in conjunction
with pontoon-bridges using wooden floats, cf id. 25.6.15 and Zos. 3.39.5. Cf the raft-
bridge described and illustrated by Apoll. Dam. Polior. 191.5 W.

‘‘Reading loris, an emendation by Stewechius. The MSS reading sociis would mean “for
their comrades”, cf III.5 fin. Perhaps they were specially trained at swimming with rafts
of arms across rivers, cf the Batavi, Tac. Agric. 18.4, Hist. 4.12.3, Cass. Dio 69.9,
M.W.C. Hassall, “Batavians and the Roman conquest of Britain”, Britannia i (1970)
131-136.

^monoxyli (Greek). Cf 11.25.

BOOK III: FIELD STRATEGY AND TACTICS

thus speedily constructed, tied together by ropes which should be kept
for the purpose, provides the solidity of a masonry arch in quick time.'

The enemy often launch rapid ambushes or raids at river crossings.
Armed guards are stationed against this danger on both banks, lest the
troops be beaten by the enemy because they are divided by the
intervening river-bed. It is safer to build stockades along the bank on
either side, and bear without loss any attack that is made. But if a
bridge is needed not just for one crossing, but for returning and for
supply-lines, broad fosses are dug around each bridgehead and a rampart
constructed to receive soldiers to defend and hold it for as long as
strategic needs require.^

8. How to lay out a camp/

It seems fit, once the disposition of a march has been described, to
move on to consider the camp in which one is to stay. For a walled city
is not always available in wartime to provide a halting-place'' or night-
quarters,^ and it is reckless and full of danger for an army to bivouac at
large without any fortification. For it is easy to contrive ambushes when
soldiers are busy taking their meal or scattered to do their duties/

‘Cf. Amm. 24.7.4 (AD 363): subiectis ignibusexuh cunctas iusserat navespraeter minores
duodecim, quas profuturaspangendispontibus disposuit vehi carpentis... “he had ordered
the burning of all the boats by fires placed under them, except twelve smaller ones to be
used for constructing bridges, which he arranged to have transported on wagons...” 24.7.8:
maerebat tamen ob haec imperator et miles, quod nec contabulandi pontis erat facultas
amissis navibus (ernere... “both Emperor and soldier were aggrieved that there was not a
chance of constructing a bridge, having lost the boats through carelessness...” Cf Zos.
3.26.3. Note that “single timber” floats were taken over from the Gauls by Hannibal at
the crossing of the river Rhone, 218 BC, cf Livy 21.26.8-9.

^The Anon, de Rebus Bellicis 16.5 stationed manuballistarii on either bank to defend
those building the bridge. Apoll. Dam. Polior. 191.5 W. attached hinged screens to his
bridge to cover those defending it.

^A shorter version of largely the same material is given in 1.21-25. See F. Lepper, S.
Frere, Trajan's Column (Gloucester 1988) Pt. 4, pp. 260-266 “Roman Camps and Forts”,
D. Baatz, “Quellen zur Bauplannung rOmischer Milit^rlager”, Mavors XI (Stuttgart 1994)
315-325.

*stativa, cf SHA Alex Sev. 45.2, where stativae (stationes?) seem virtually identical to
mansiones.

^mansio

^'munera.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

moreover the darkness of night, the need for sleep and the dispersal of
grazing horses all provide opportunities for attack.

When surveying a camp, it is not sufficient to choose a good site
unless it be so good that no other site better than it can be found.
Otherwise a more advantageous site overlooked by us may then be
occupied by the enemy, bringing danger. Also ensure that unhealthy
water is not close by nor wholesome water too far away in summer, that
there is no shortage of fodder and firewood in winter, that the site on
which one is to camp is not liable to flooding after sudden rainstorms,
that it is not in broken, remote country where the enemy may surround
us and make it difficult to escape, and that missiles cannot be shot from
higher ground by the enemy so as to reach it.^

When these conditions have been carefully and stringently
investigated, you may build the camp square, circular, triangular or
oblong, as required by the site. Appearance should not prejudice utility,
although those whose length is one-third longer than the width are
deemed more attractive. But surveyors^ should calculate the square
footage of the site-plan^ so that the area enclosed corresponds to the size
of the army. Cramped quarters constrict the defenders, whilst unsuitably
wide spaces spread them thinly

There are potentially three sorts of fortification defined for a camp.
The first is for the passage of one night or for brief occupation on a
march. The raised turves are laid out in line, forming a rampart.^ Above
it, va///,^ that is, stakes’ or wooden spars® are ranged along its length.
The turf is cut around with iron tools, retaining the earth in the grass
roots, Yi ft. high, 1 ft, wide and 1 V 2 ft. long. When the earth is too loose

'Cf. Ps.-Hyginus 57.

^aghmensores.

^podismus mensurae.

'‘Cf. 1.26, in.6. The free shape of the Roman camp appears late-antique, cf 111.10, where
V. recommends the Gothic (circular) carrago or “wagon-city”. Contrast Ps.-Hyg. 21,
Polyb. 6.31.10, 6.42.

^agger.

^'stockades.

^sudes.

'^tribuli.

BOOK III: FIELD STRATEGY AND TACTICS

for it to be possible to cut out the turf like a brick, the fosse is dug in
“temporary style”,' 5 ft. wide, 3 ft. deep, with the rampart rising on the
inside. Thus the army is enabled to rest secure and without fear.

But a stationary camp^ is fortified with greater care and effort,
whether in summer or winter, when the enemy is near. For each century
receives a footage^ apportioned by the drillmasters and officers.'' The
men distribute their shields and packs in a circle around their own
standards and, armed only with a sword, open a fosse 9 ft. wide, or 11
ft. or 13 ft. or, if a major hostile force is feared, 17 ft.—it is usual to
keep to odd numbers. The rampart is then raised between lines of
revetments or barriers of logs and branches interposed to stop the earth
easily falling away. Above it a system of battlements and turrets is
constructed like a wall. The centurions measure the work with ten-foot
rods, checking that no one’s laziness has resulted in digging too little or
making mistakes. The tribunes also go round and, if they are conscien¬
tious, do not go away until it is completed in every part. However, to
prevent a raid from being mounted on the men at work, all cavalry and
that part of the infantry which through the privilege of rank does not
labour take up position in front of the fosse in an armed cordon and re¬
pel enemy attack.

So first the standards are set up in their places inside the camp,
because nothing is more revered by the soldiers than their majesty, the
headquarters^ is prepared for the general and his staff-officers^ and the
pavilions^ are erected for the tribunes, who are served with water,
firewood and fodder by privates* assigned to services."^ Next, in order of

'opere tumultuario, cf. 1.24, where it is 9 ft. wide and 7 ft. deep. Obviously the first sort
of fortification has regard only for the turf rampart, the second, “emergency work/fosse”
and third, “proper work/fosse” for the rampart and fosse as a defensive system of
differential sizes.

^stativa castra.

^pedatitra, cf Ps.-Hyginus Castr. 1, etc.

^campidoctores et principia,, here substituted for centurions and tribunes mentioned
below; cf 111.6.

^praetohum.

^'comites, “counts”.

^ tabernacula.

^contubernales.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

rank, sites are appointed in camp for the legions and auxiliaj for
cavalry and infantry, to pitch their tents.^

From each century four cavalrymen and four infantrymen undertake
sentry-duty by night.^ Because it was clearly impossible for individuals
to remain constantly awake in their look-out posts, the night-watches'*
have been divided into quarters by the water-clock, ensuring that is
necessary to be awake for no more than three hours a night.^ All the
watches are called by the trumpeter and at the end of their time recalled
by the homblower.^ The tribunes select able and very reliable men to
patrol the watches and report any fault that emerges. These used to be
called circumitores; they have now been made a rank of service and are
called circitoresJ Note that the cavalry should do night watch-duties
outside the rampart. During the day in the case of a stationary camp*
they change guard (merely) for morning and afternoon shifts in the out-
stations,^ to avoid exhausting the men and horses.

Among the things particularly incumbent upon a general, whether he
is quartered in a camp or a city, is to see that the animals’ pasturage,

'^munera.

'palatine auxiUa
^papiliones.

^excubitus. V.’s usage of technical terminology is often non-technical, cf Is id. Etym.
9.3.42: excubiae autem diurnae sunt, vigiliae nocturnae^ '^'Excubiae are daytime, vigiliae
are night-time.” However, cf. Caes. Bell. Gall. 7.69.7: haec eadem noctu excubitohbus
ac flrmispraesidiis tenebantur, “The same (fortlets) were held at night by excubitores and
strong garrisons.”

^vigiliae.

^Cf Aen. Tact. 22.24-25, Onas. 10.10-12, Ps.-Hyg. 1, Polyb. 6.35.1-5, ?\i\\o In Flaccum
3, J. Marquardt, Romische Staatsverwaltung II (1885^) 420.

^’This conflicts with the signalling convention described in 11.22, and the bucina was long
identified with the watch-signal (and watch-period); cf. J. Marquardt, 421.

’“patrolmen”. Of circumitores, Petronius 53.10, of a civilian, is the only attestation. Here
it is perhaps V.’s etymologizing invention to explain circitores. Cf. Sail. Jug. 45.2, Polyb.
6.35.8-12, 6.36.1-9, Tac. Hist. 2.29, Ann. 15.30, Jos. Bell. lud. 5.510. V.’s comment on
circitores is confirmed by Fink, doc. 47.i.7 and 17, doc. 49.2. It was the legal duty of a
commander clavas portarum suscipere vigilias interdum circumire, “to hold the keys to
the gates and patrol the watches regularly”, cf Dig. 49.16.12.2. See R.W. Davies,
“Minucius lustus and a Roman Military Document from Egypt”, Aegyptus liii (1973)
75-92, esp. 88-90.

^castra posita.

‘^agrariae.

"’Cf Livy 44.33.10-11 (168 BC), an innovation by L. Aemilius Paullus. Cf III.22 note.

BOOK III: FIELD STRATEGY AND TACTICS

the transportation of grain and other provisions, and the ministration of
water, firewood and fodder are rendered secure from hostile attack.' The
only way to achieve this is to plant garrisons at suitable points through
which our supply trains pass. These may be cities or walled forts. If no
old fortifications are available, temporary forts are established in
favourable positions and girded with broad fosses; such castella^ are
named after castra^ by a diminutive word. A number of infantry and
cavalry stationed in them on outpost-duty provide a safe passage for
supplies. The enemy hardly dares attack points where he knows his
adversaries are camped ahead and behind.'*

(Pre-battle strategy, ch. 9-13)

9. What and how many things are to be considered when judging
whether to engage the enemy in raids and ambushes or else in
pitched battle.

Whoever will deign to read these commentaries on the art of war
abridged from authors of the highest repute, wishes to hear first and
foremost the science of battle and the recommended tactics. But a
pitched battle is defined by a struggle lasting two or three hours, after
which all hopes of the defeated party fall away. That being so, every
expedient must be thought of previously, tried out in advance and
implemented before matters come to this final pass. For good generals
do not attack in open battle where the danger is mutual, but do it always
from a hidden position, so as to kill or at least terrorize the enemy while

'On the duty of a commanding officer to ensure an adequate supply of food, see R.W.
Davies, “Some notes on military papyri from Dura”, BASF v (1968) 33 and nn.15-16.
^forts. A smaller size of castella are defined as burgi at IV. 10.

Tortress-camp.

‘‘The fortification of roads with burgi is attested in late-fourth-century Britain, Belgium,
Spain and Pannonia; cf. R. MacMullen, Corruption and the decline of Rome (Yale 1988)
187.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

their own men are unharmed as far as possible. In this connexion I shall
describe the measures which the ancients found quite essential.'

An important art useful to a general is to call in persons from the
entire army who are knowledgeable about war and aware of their own
and the enemy’s forces, and to hold frequent discussions with them in
an atmosphere from which all flattery, which does so much harm, has
been banished, to decide whether he or the enemy has the greater
number of fighters, whether his own men or the enemy’s are better
armed and armoured and which side is the more highly trained or the
braver in warfare.^ A further question is which side has the better
cavalry or infantry, bearing in mind that the strength of an army
depends mainly on its infantry.^ And, among the cavalry, which side has
more pikemen or archers, which is wearing more cuirasses and which
has brought better horses. Then he should consider whether the terrain
itself in which one is to fight appears advantageous to the enemy or to
ourselves. For if we are strong in cavalry, we should opt for plains; if
in infantry, we should choose confined places, obstructed by ditches,
marshes or trees, and sometimes mountainous."* Also, which side has
more food or lacks it, for hunger, they say, fights from within, and
often conquers without a blow.^

But most important of all, he should deliberate whether it is expedient
for the crisis to be prolonged or fought out more swiftly. For sometimes
the enemy hopes that the campaign can be ended quickly, and if it
becomes long-drawn out, is either reduced by hunger, or called back to
his own country by his men’s homesickness, or through doing nothing
significant is compelled to leave in despair. Then very many desert,

‘V. was writing in the aftermath of a disastrous pitched battle with the barbarians
(Adrianople, AD 378), and has a correspondingly high opinion of covert or guerilla
actions. It seems from this chapter that he considered the Romans inferior in numbers and
strength to the barbarians and did not trust the loyalty of the auxilia palatina. Cf 11.3,
where he contrasts their looser discipline and readier privileges with the legions.

^Cf. 1.2 fm.. Onas. 3.1-3, 11.6, and Amm. 31.12.7 on the damage done by adulation in
council.

'Cf. II. 1.

'Cf. III.13,

'Cf. II1.3.

BOOK III: FIELD STRATEGY AND TACTICS

exhausted by effort and weariness, some betray others and some
surrender themselves, since loyalty is less common in adversity, and the
enemy who came in great force begins to be denuded.*

It is also relevant to find out the character of the adversary himself,
his senior staff-officers^ and chieftains.^ Are they rash or cautious, bold
or timid, skilled in the art of war or fighting from experience or
haphazardly? Which tribes on their side are brave or cowardly? What
is the loyalty and courage of our auxilial"^ What is the morale of the
enemy forces? What is that of our own army? Which side promises
itself victory more?^ By such considerations is the army’s courage
bolstered or undermined.

When the men despair, their courage is raised by an address from the
genera], and if he appears fearless himself, their spirits are raised, if for
example you have brought off some exploit from an ambush or
opportunity, if the opposition have begun to suffer mishaps, or if you
have been able to overcome some of the weaker or poorly-armed
elements of the enemy.^ Be careful never to lead a hesitant and
frightened army into a pitched battle. It matters whether you have an
army of recruits or veteran soldiers, and whether they were on active
service a short time before or have spent a number of years at peace.
For men who stopped fighting a long time ago should be treated as
recruits.

Indeed, when legions, auxilia' and cavalry^ arrive from different
stations, the best general should have them trained by picked tribunes
of known conscientiousness in all types of arms separately as single
units, and after forming them into one body, he will often train them

'Cf Claud, pan. de Hon. cons. VI 250-264, Alaric, defeated by Stilicho at Pollentia AD
402-403, suffers a wave of desertions.

^ comites.

^optimates.

^t.e. palatine auxilia.

^Repromitto, cf relego II.4, with late-Latin redundant prefixes.

'Cf III. 12.

^i.e. palatine auxilia.

“i.e. vexillationes^ cf III. 10.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

himself as if for fighting a pitched battle, and will test them to see what
their potential skill and courage may be, how well they interact with one
another and whether they obey promptly the warnings of trumpets,
directions of signals and his own orders and authority. If they err in any
respect, let them be trained and instructed for as long as it takes to
become perfect. But if they become fully expert in field manoeuvres,'
archery, throwing javelins and drawing up the line, they should not even
then be lightly led into a pitched battle, but on a carefully chosen
opportunity, and only after being blooded in smaller-scale conflicts.

So let the general be watchful, sober and discreet.^ Let him call a
council-of-war and Judge between his own and the enemy’s forces, as
if he were to adjudicate between parties to a civil suit. If he finds
himself superior in many particulars, let him be not slow to enter a
battle favourable to himself If he recognizes that the enemy is stronger,
let him avoid a pitched battle, because forces fewer in number and
inferior in strength carrying out raids and ambushes under good generals
have often brought back a victory.

10. What to do if one has an army unaccustomed to fighting or
newly recruited.

All arts and all works progress through daily practice and continual
exercise. If this is true of small things, the principle should hold all the
more true in great matters. Who can doubt that the art of war comes
before everything else, when it preserves our liberty and prestige,
extends the provinces and saves the Empire? The Spartans long ago
abandoned all other fields of learning to cultivate this, and later so did
the Romans.^ Even today the barbarians think this art alone deserves
their attention; they are sure that everything else either depends on this
art or can be obtained by them through it."* It is essential to those whose

'campicursio.

'Cf. Onas. 1.1.

'Cf. Ill praef.

'Cf. Sen. deJra 1.11.3-4.

BOOK III; FIELD STRATEGY AND TACTICS

business is war, for it is the means to hold on to life and win a victory.*
So the general who has bestowed on him the insignia of great power,
and to whose loyalty and strength are entrusted the wealth of
landowners, the protection of cities, the lives of soldiers and the glory
of the State, should be anxious for the welfare not just of his entire
army, but for each and every common soldier also." For if anything
happens to them in war, it is seen as his fault and the nation’s loss.^
Therefore if he is leading an army of recruits or of men long
unaccustomed to bearing arms, let him thoroughly explore the strength
and spirit of each legion, auxilium^ and vexillatio.^ Let him find out by
name if possible the military potential of each “count”/’ tribune,^ aide*
and private.^ Let him assume maximum authority and severity, punish
all military crimes according to the laws, have a reputation for forgiving
no errors and make trial of everyone in different places in diverse
situations.When he has seen to these things properly, let him choose

'Cf. 11.24.

^Cf 1.7

^Cf. SHA Alex. Sev. 47,1: dicens milites se magis servare quant se ipsum, quod salus
publica in his esset, “saying that he preserved the soldiers more than his own person,
because the national salvation rested on them.”

'^palatine auxiliary regiment.

^cavalry vexillation.

^'comes rei militaris or regimental commander.

’Also a regimental commander, properly a comes, but often the vicarius seems to have
stood in for his boss as “tribune”, usurping his title, cf Jones 1278, n, 158; V. calls him
the “lesser tribune” (II.7).

^domesticus, cf. Grosse, 120 ff. The domesticus, like the vicarius, was usually a
primicerius, whose rank varied according as his boss’s. V. likens his centurio primi pili
to ?L primiscrinius = primicerius (II.21). As bureau-chief, the domesticus was often called
dtprinceps (11.12); V. has difficulties visualizing an “ancient legion” having both 2 iprimus
pilus and a princeps (II.8).

'^contubernalis.

'^Cf. M. Pronto ad Verum Imperatorem Aurelium Caesarem II. 1.23 = Cato Dierum
Dictarum de consulatosuo fr. 13 (Jordan p.35): Interea unamquqmque turmam manipulum
cohortem temptabam, quidfacere possent; proeliis levibus spectabam cuius modi quisque
esset; si quis strenue fecerat, donabem honeste, ut alii idem vellent facere, atque in
contione verbis mult is laudabam. Interea aliquot <p>au<ca> castra feci, sed ubi anni
tempus venit, castra hiberna... “Meanwhile I tested each and every troop, maniple and
cohort, to see their potential. In small-scale encounters I observed each man’s
characteristics; if anyone acted bravely I rewarded him handsomely, so that others would
wish to do the same, and I praised him at length in addresses. Meanwhile I constructed
a number of small camps, and when the season of year came, a winter camp...”

VEGETIUS’ EPITOME OF MILITARY SCIENCE

a moment when the enemy are roving carelessly about, scattered for
ravaging, to send in his well-tried cavalry or infantry accompanied by
the new recruits or poor-quality soldiers. Routing the enemy at a
favourable opportunity gives experience to the latter group, and raises
the morale of the rest.

Let him set up ambushes in complete secrecy at river-crossings,
mountain passes, wooded defiles, marshes and other difficult passages.
Let him so regulate his march that, fully prepared, he attacks the enemy
when they are suspecting nothing, when they are eating meals, sleeping
or at any rate resting, when they are relaxed, unarmed, unshod and their
horses unsaddled, to the end that his men may acquire self-confidence
in battles of this kind. For those who have not for a long time, or never
at all, seen men being wounded or killed are greatly shocked when they
first catch sight of it, and confused by panic start thinking of flight
instead of fighting. Also, if the enemy are ranging abroad, let him attack
them when they are fatigued by a long march; let him harass the rear,
or at least attack by surprise, and those who loiter at a distance from
their people for fodder or plunder, let him attack suddenly with picked
men. Those actions should be tried first which do less harm if they fail,
and bring the most benefit if successful.'

It is (also) the mark of a skilled general to sow seeds of discord
among the enemy. For no nation, however small, can be completely
destroyed by its enemies, unless it devours itself by its own feuding.
Civil strife is quick to compass the destruction of political enemies, but
careless about the readiness of (the nation’s) own defence.

There is but one premiss to this work: let none despair of the
possibility of doing that which was done in the past.^ One may say, it

'This strategy is illustrated by Sebastianus’ raid on the marauding Goths on the eve of
Adrianople, ad 378, in striking contrast to Valens’ action in forcing a pitched battle. Cf.
Amm. 31.10.4, and the historian’s approval of guerilla tactics against the Goths, Amm.
31.7.2. Similarly Modares, a Gothic magister militum in Thrace AD 380-382, inflicted a
crushing defeat on a Gothic band caught in the act of enjoying their plunder, cf. Zos.
4.25.2-4.

^V. reverts to the theme of 1.21 ff. on the failure to build camps and the disastrous results
thereof. There is some suggestion that Theodosius I’s field armies did not always build
camps on campaign when it was clearly advisable to do so, cf. Zos. 4.31.3-4, 4.49.

BOOK III: FIELD STRATEGY AND TACTICS

is many years since anyone enclosed an army encampment with a fosse,
rampart and stockade. We shall respond, if that precaution had been
taken, there is nothing that the enemy could have done to harm us by
attacking by night or day. The Persians copy the Romans in building
their camps with lines of fosses. And because almost all of them are in
sandy areas, they build their rampart using sacks which they have
brought empty with them, filling them up with the dusty earth dug out
of the fosse, and forming them into a pile.* All barbarians spend nights
secure from attack behind their wagons linked together in a circle like
a military camp.^ Are we afraid that we are unable to learn what others
have learned from us?

These skills were formerly maintained in use, as well as in books, but
once they were abandoned it was a long time before anyone needed
them, because with the flourishing of peacetime pursuits the imperatives
of war were far removed.^ But lest it be thought impossible for an art
to be revived whose use has been lost, let us be instructed by
precedents. Among the ancients, military science often fell into oblivion,
but at first it was recovered from books, and later consolidated by the
authority of generals. Scipio Africanus took over our armies in Spain
after they had been several times beaten under other commanders. By
observing the rule of discipline, he trained these so thoroughly in every
article of work and digging of fosses, that he said that they deserved to
be stained by digging mud, because they had declined to be wetted by
the enemy’s blood."' With these same men, he eventually captured the

'Cf. baskets filled with sand and used to build temporary fortifications in Aen. Tact. 32.1,
8 .

^The carrago or “wagon-city” of the Goths, Huns and Alans; cf Amm. 31 7.7, cf. 7.5
(Goths), 31.2.18 (Alans, cf. Arr. Anab. 5.22.4, from autopsy?), Amm. 31.2.10 (Huns).

'Cf. 1.28.

'’So Hlor. 1.34.10: quippe adsiduis et miustis et servilibus maxime operibus adtriti ferre
plenius vallum, qui arma nescirent, luto inquinare, quia sanguine nollent, iubebantur, “for
worn out by continual, excessive and above all servile work, they were ordered all the
more to carry stakes, since they did not know how to carry arms, and to be stained by
mud, since they refused blood.” Cf 1.20 ff on the training-value of building
encampments. The sixth-century Maurice was also credited with having restored this
discipline to the army, after it had been forgotten, cf Men. Prot. fr. 23.3 (Blockley).

VEGETIUS’ EPITOME OF MILITARY SCIENCE

city of Numantia, and so cremated the inhabitants that none escaped.*
Metellus took over an army in Africa which had
been sent under the yoke when Albinus was its commander. He
reformed it on ancient principles, and later overcame the same men who
had sent them under the yoke.^ The Cimbri destroyed the legions of
Caepio and Mallius inside Gaul. The remnants were taken up by Gaius
Marius, who trained them in the knowledge and art of warfare. The
result was that they not only destroyed an innumerable host of Cimbri,
but of Teutones and Ambrones as well, in a general engagement/ But
it remains easier to train new men in valour than to reanimate those who
have been terrified out of their wits.^

11. Precautions to be taken on the day of engaging in a general
action.^

After treating ot ihc lesser skills of war, our analysis of miiitary science
invites us to consider the hazard of the general engagement, the fateful
day for nations and peoples. For total victory depends upon the outcome
of an open battle. Therefore this is the time when generals should exert
themselves all the more, in proportion as the vigorous may hope for
greater glory, and worse peril dogs the slack. This is the moment when
application of skill, theory of warfare and planning dominate.^

'P. Cornelius Scipio Aemilianus (Africanus) (Numantinus), cos. 147 bc, defeated and
destroyed Carthage in 146, took over command of the campaign at Numantia, Spain, as
consul 11 in 134, retrained the army, and captured and destroyed the city in 133 BC.

^Q. Caecilius Metellus (Numidicus), cos. 109, took over command in the war against
Jugurtha after the defeat of the army under A. Postumius Albinus, legatus pro praetore,
in 110 BC. Sail. Jug. 44-45 tells the story of his retraining programme.

’C. Marius defeated the Cimbri at Vercellae in 101 bc and the Teutones and Ambrones
at Aquae Sextiae in 102 bc avenging the defeat of the Roman armies of Q. Servilius
Caepio, cos. 106, and Cn. Mallius Maximus, cos. 105, in 105 bc, and other defeats. These
examples presuppose that V. was still writing in the aftermath of military defeat (cf. 1.20),
despite the victories praised in II praef

'Cf. 1.13 fin.

^The material on the battle (111,11-22) is stylistically distinct. The frequent use of the
verbal 2nd pers. sing, and of tuus-a-um is marked; cf. D. Schenk, 60-61. There are other
grounds (cf. 111.17,111.20) for thinking parts of the section are based largely on Cato’s lost
de Re Military and this appears to be the best explanation of the style, too.

‘’Cf. Colum, 1.1.16: usus et experientia dominantur in artibus, “method and experience
in ^e arts and sciences dominate.”

BOOK III: FIELD STRATEGY AND TACTICS

In ancient times it was customary to lead soldiers into battle after
they had been treated to a light meal, so that the food ingested might
give them strength,* and in a prolonged fight they might not grow tired
from hunger.^ You should also take care if you lead your men to battle
from a camp or a city when the enemy are present, lest, while the army
is marching out in defile through narrow gates, it may be worsted by
massed and prepared hostile forces. Therefore one should ensure that all
soldiers get clear of the gates and form a battle-line before the enemy
arrives. But if he arrives prepared for battle while your men are still
inside the city, postpone your exit or at least pretend to. Then when the
enemy troops start hurling insults at men they do not expect to come
out, when they turn their attention to booty or withdrawal, when they
break ranks, that is the moment for your crack troops to sally forth
against the stunned enemy and attack them in force unexpectedly.^
Beware also not to force to a pitched battle soldiers who are tired
after a long march or horses that are weary from galloping. Men who
are going to battle lose much of their strength from marching-fatigue.
What is one to do, if he reaches the line exhausted?"* This is something
the ancients avoided, and in the recent past^ it was the armies, to say no
more, who learned the lesson after Roman generals had through lack of
expertise failed to provide against it.^ For when a tired man enters battle
with one who has rested, or a sweating man with an alert, or one who
has been running with one who has been standing, he fights on unequal
terms.

'Cf. Onas. 12.

^Cf. Front. Strut, 2.1.1, 2.1.5, and Belisarius’ defeat with a hungry army at Callinicum
AD 531, Procop. Bell, Pers. 1.18.13 ff.

^Cf Onas. 10.20.

‘‘Cf Onas. 6.9.

^ super lore vel nostra aetate, cf. 1.28.

reference to the battle of Adrianople, AD 378, where the Emperor Valens was killed
along with two-thirds of the Eastern field army by the Goths, Huns and Alans,
commanded by Fritigem; cf. Goffart, 83, and Sabbah, 142. V. uses hyperbaton to
emphasize the words hoc... exercitus didicerunv, cf. 1.10 fm. for a similar figure in a
possible allusion to the Tigris disaster, ad 363. Amm. 31.12.10-13.7 stresses that the
battle followed upon an exhausting forced march in the heat of the day from Adrianople
itself to the Gothic camp, where the battle took place, the troops not having been fed or
watered. At 31.14.5 he says that Valens was skilled in neither military nor liberal studies.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

12. One should find out how soldiers are feeling before battle.
Explore carefully how soldiers are feeling on the actual day they are
going to fight. For confidence or fear may be discerned from their facial
expression, language, gait and gestures. Do not be fully confident if it
is the recruits who want battle, for war is sweet to the inexperienced.’
You will know to postpone it if the experienced warriors are afraid of
fighting. An army gains courage and fighting spirit from advice and
encouragement from their general, especially if they are given such an
account of the coming battle as leads them to believe they will easily
win a victory. Then is the time to point out to them the cowardice and
mistakes of their opponents, and remind them of any occasion on which
they have been beaten by us in the past. Also say anything by which the
soldiers’ minds may be provoked to hatred of their adversaries by
arousing their anger and indignation.^

It is a natural reaction in the minds of nearly all men to be fearful as
they go to do battle with the enemy. But those whose minds are
panicked by his actual appearance are without doubt the weaker sort.
Their fears may be lessened by the following remedy. Before the battle,
repeatedly draw up your army in safe positions from which they can get
used to seeing and recognizing the enemy. Let them also try their hand
now and then when an opportunity arises: let them put to flight or kill
their opponents; let them learn to recognize their adversaries’
characteristics, arms and horses,^ for familiar things are not frightening."*

13. How a suitable place is chosen for battle.

The good general should know that a large part of a victory depends on
the actual place in which the battle is fought. Be at pains therefore when

'Cf. Pindar fr. 110: yXuKu 6e TtoXe^icx; dTceipomv, e|i7t8ipcov 6e Tiq | rappeC
Ttpoaid via viv KapSiQt TtEpiaacS <;. “War is sweetto the inexperienced, butan experienced
man fears it as it approaches with all his heart, extraordinarily.”

^Cf. III.9. Aeneas Tacticus appears to have written a whole book— Akousmata —on things
to say to the troops, cf. Aen. 38.5. Cf. also Caes. Bell Gall 1.40.4 ff.

'Cf. III. 10. Cf the strategy of Q. Fabius Maximus (Cunctator) for the demoralized
survivors of the battle of Lake Trasimene in 217 BC, Livy 22.12.8-10.

^Cf Plut. Mar. 16. V. is unaware of Tac. Agr. 30.4: omne ignotum pro magnifico est.
“Every unknown is assumed to be great.”

BOOK III: FIELD STRATEGY AND TACTICS

you are going to engage in combat, to get help first from the place. This
is judged the more advantageous, the higher the ground occupied. For
weapons descend with more violence onto men on a lower level, and the
side which is higher dislodges those opposing them in greater force. He
who struggles uphill enters a double contest with the ground and with
the enemy. But there is this distinction: if you are hoping for victory
from your infantry over enemy horse, you should choose rough, broken
and mountainous country; but if you are looking for victory from your
cavalry over opposing infantry, you should go for positions that are,
indeed, on a slightly higher level, but flat and open, unobstructed by
woodland or morasses.^

(Battle tactics and strategies, ch. 14-26)

14. How the line should be drawn up to render it invincible in
battle.^

When the general is ready to draw up his line, he should attend first to
three things, sun, dust and wind. When the sun is in front of your face,
it deprives you of sight. Head-winds deflect and depress your missiles,
while aiding the enemy’s. Dust thrown up in front of you fills and
closes your eyes. Even inexperienced generals usually avoid these things
at the time of ordering the lines, but the provident general should take
care of the future lest, a little while later as the day wears on, the
changed position of the sun may be harmful or a head-wind may
habitually arise at a regular time, during the fighting. Therefore let the
lines be ranged with these problems behind our backs, and if possible
so that they may strike the faces of the enemy.^

‘cf. m.9.

^Essentially the same material as 11.15-17 (and assumed in 1.20), except that the light
lines are here differentiated by their own ordinal number. The material is based on Cato;
cf Livy 1.43, 8.8, Dion. Hal. 4.16, for similar arrays. Developments under the Principate
did not render the basic patterns obsolete, cf Arr. Ect. 12-26, Amm. 14.6.17.

^This was above all the problem to which the Romans attributed their defeat by Hannibal
at Cannae in 216 BC. Cf Flor. 1.22.16: callidus imperator (sc. Hannibal) in patentibus
campis observato loci ingenio, quod et sol ibi acerrimus et plurimum pulvis et eurus ab
oriente semper quasi ex constituto, ita instruxit aciem, ut, Romanis adversus haec omnia
obversis, secundum caelum tenens vento pulvere et sole pugnaret, “The cunning

VEGETIUS’ EPITOME OF MILITARY SCIENCE

“Line”* means the army drawn up for battle. The “front”^ is the part
that looks towards the enemy. If wisely deployed it is very useful in a
general engagement. If unskilfully, however excellent the warriors may
be they are weakened by bad ordering. The rule of drawing up an array
is to place in the first line the experienced and seasoned soldiers,
formerly called principes, and to rank in the second line archers
protected with cataphracts and crack soldiers armed with javelins^ and
light spears,"* formerly called hastati? Individual infantrymen regularly
occupy 3 ft. each. Therefore in a mile^ 1,666 infantrymen are ranked
abreast, without light showing between them but leaving room to handle
their weapons. Between line and line, they wished to have a space 6 ft.
in depth behind them to give the fighting men room to move forward
and back, missiles being more forcibly thrown from a running jump.’
In these two lines are posted those older in years, confident and
experienced, and protected by heavy armour. Their role is to act like a
wall;® at no time should they be made to retreat or pursue lest they
disturb their ranks. They should receive oncoming adversaries and repel
or rout them by standing their ground and battling it out.

commander-in-chief (Hannibal) observed the character of the place, that both the sun was
very fierce there and there was much dust, and that the east wind blew always from the
east as if by covenant, so he drew up his battle-line in the spreading plain in such a way,
that the Romans might be facing all these disadvantages while he himself, keeping the sky
in his favour, would be fighting along with the wind, dust and sun.” Cf 1.38.15, Front.
Strat. 2.2.7. Cf. Aug. c/v. Dei 5.26 D, Oros. 7.35,17-18 for a “miraculous wind” helping
Theodosius 1 against Eugenius at the battle of the river Frigidus, AD 394.

‘flc/ej, literally “cutting edge”.

Efrons.

^spicula^ cf. 11.15.

*lanceae.

^V. intends to follow his source in that the first and second line were heavy-armed, but
the contemporary cataphract archers and spearmen he substitutes appear to have been
something in between the old heavy and light categories, cf Procopius cited at 11.15.

^’1 Roman mile = c.1,617 yds., 1,478*5 m., cf 1.9 note.

^Cf Polyb. 18.30.6-7, who suggests the more individualistic style of combat used by the
Roman lines required minima of 3 ft. between man and man in both directions. V.
wrongly assumes no space at all laterally, as in the Macedonian phalanx, because this is
incompatible with the generous forward-and-back movement he also allows.

*Cf 11.17, 1.20.

BOOK III: FIELD STRATEGY AND TACTICS

The third line is formed from very fast light infantry,‘ young archers
and good javelinmen. They were formerly termed ferentariir The fourth
line is similarly constructed from very light ' shield-bearers”,^ young
archers and those who fight briskly with light javelins'* and lead-
weighted darts called plumbatae.^ They used to he termed “light
armament”.^ It should be noted that whereas the from two lines stand
their ground, the third and fourth lines always go out to challenge the
enemy with missiles and arrows, in the forward position. If they can put
the enemy to rout, they set off in pursuit themselves with the cavalry.
But if they are driven bv the enemy, they return to the first and
second lines and retire between them to their own stations.^ The first

and second lines bear the full brunt of the battle when it comes to what
is called “to broadswords and javelins”.^

In a fifth line were sonn nes placed carriage-ballistas,'^ and

manuballistarii,' ' “sling-staff men”“ and slingers. “Sling-staff men” are
those who cast stones from a “sling-staff’.'^ The “sling-staff’ is a staffs

^armaturae.

^Used by Cato fr. 6 de Re Mil. and Sallust Cat. 60.2, archaic by the time of Livy and
Caesar. Cf. Varro apud Non. Marc. 520.10 M.: vocabant ferentarios, qui depugnabant
fundis et lapidibus, “They called ferentarii {hose who fought with slings and stones.” Paul.
Test. 506.26 L; Cato eos ferentarios dixit, qui tela ac potiones militibus proeliantibus
ministrant. “Cato called them ferentarii, who served weapons and drinks to the soldiers
as they fought.”

^scutati^ i.e. again probably late-Roman archers or spearmen with shields attached to their
body-armour. Cf. Procop. Bell. 1.1.8-16 for similar horse-archers in the early Byzantine
army; cf. at 11.15 V.’s identification of these light scutati with modern soldiers.

Vrwta, cf 11.15.

^Variatio for plumbatae called mattiobarbuli, cf 1.17, showing the possible extent of
manipulation of technical terms for stylistic effect.

^'levis armatura.

"Cf 1.20, 11.17.

'^ad spathas et ad pila. Obviously the close-quarters phase of battle is intended. The
technical phrase is not otherwise attested, however. Cf 1.20 fin. where it is called ad pila.

'^carroballistae. Cf Arrian’s array, Ect 19, for the positioning of artillery behind all the
lines and on the flanks. Cf also 11.15, 11.25, III.24.

“’crossbowmen or catapultiers.

' ^fundibulatores.

^^fustibalus.

^^Justis.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

4 ft. long, attached to the middle of which is a sling’
made of leather, and operated with either hand, it discharges stones
almost like the mangonel.^ Slingers^ discharge stones from slings made
of flax or hair—the latter are said to be better—by whirling the arm
about the head. Those not issued with shields fought in this line,
throwing stones and missiles by hand."* They were called “extras”, as
being recruits, added afterwards.^

The sixth line behind all the others was held by very reliable
warriors, armed with shields and every type of arms. These the ancients
called triarii. They would wait in reserve behind the last lines, to attack
the enemy more violently, being themselves rested and intact. If
anything happened to the lines in front, all hope of recovery depended
on their bravery.^

15. Counting the square-footage, how much space in the line should
be left abreast between each man, and in depth between each rank.
Now that it has been explained how the lines should be drawn up, I
shall discuss the square footage^ and plan of the formation itself. In a
mile of field,* a single line will contain 1,666 infantry, since individual
fighting men take up 3 ft. each.^ If you wish to draw up six lines in a
mile of field, 9,996 infantry are needed. If you wish to deploy this

^Junda.

^onager, “wild ass”. The “sling-staff’ is known from mediaeval art also, cf. T.G. Kolias,
Byzantinischen Waffen (Vienna 1988) 255, with refs. The sling was actually attached to
the top end, not the middle as V. says. At IV.22, q.v., V, refuses to describe the
fustibalus, arcuballista and Jitnda on the grounds of their familiarity, evidently forgetting
that he had described two of them here.

^funditores.

'‘Cf. 1.16, 11.23. Cf also Arr. Eel 25 on throwing stones.

^accensi, cf 11.19. They belong in the third and second centuries BC, but they are likened
to the modem category of tiro or “recruit”, which was a rank in itself at the bottom of the
pay-scale, cf Anon, de Rebus Bellicis 5.7 and Hieron. c. loh Hierosol. 19.

"Cf 1.20, 11.16.

^podismus.

*1 Roman mile, consisting of 1,000 Roman paces or 5,000 Roman feet, c. 1,617 yds.,
1,478*5 m., cf 1.9 note.

'"Cf 111.14.

BOOK 111: FIELD STRATEGY AND TACTICS

number in three lines, it takes up two miles; but it is better to make
additional lines than to thin the soldiers out.

We said that 6 ft. ought to lie between each line in depth from the
rear,' and in fact each warrior occupies 1 ft. standing still. Therefore, if
you draw up six lines, an army of 10,000 men will take up 42 ft. in
depth and a mile in breadth. [If you decide to draw up three lines, an
army of 10,000 will take up 21 ft. in depth and two miles in breadth.]^
In accordance with this system, it will be possible to draw up even
20,000 or 30,000 infantry without the slightest difficulty, if you follow
the square footage for the size.^ The general does not go wrong when
he knows what space can hold how many fighting men.

They say that if the field is too narrow, or if numbers are sufficient,
the lines can be drawn up ten-deep or more. For it is more useful that
they should fight in close order than too far separated. If the line is too
thinly deployed, it is quickly broken through when the enemy make an
assault, and after that there can be no remedy.'' The units which should
be deployed on the right wing, the left and in the middle, are either
distributed according to their traditional ranking, or else changed to
match the ability of the enemy.

'Cf. III.14.

^This sentence is included in MSS nV and recentiores, but was rejected as an
interpolation by Stewechius, followed by Lang. There seems no obvious reason to make
up such a sentence, which describes a valid permutation of V.’s system, and so is
probably genuine.

^Cf. the same man-space mentality for the size of the camp, 1.22, 111.8. For armies of
these sizes, cf. III.l. V.’s figures are over-generous in depth, allowing 6+1 i\. for each
line, as compared with Polybius’ 3(+l?) ft., and too mean in breadth, allowing 3+0 ft. for
each man, as compared with Polybius’ 3(+3?) ft., thus no room to move at all, cf. 111.14
note.

"Cf. Onas. 21.1-2.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

16. On deploying cavalry.*

When the infantry line has been formed up, the cavalry are posted on
the wings, so that all cuirassiers^ and pikemen^ are next to the infantry,
and mounted archers and those not issued with cuirasses range farther
afield. For the heavy cavalry' should be used to protect the infantry’s
flanks, while the swift light cavair>' are for overwhelming and throwing
into disorder the enemy’s wings.

The general should know against which drungi,"^ that is, “groups"' of
the enemy he should set which of the cavalry. For some obscure, or
indeed, one might say, divine reason, some men fight better against
others, and those who have conquered the stronger are often themselves
defeated by the weaker.^ But if the cavalry are outnumbered, the ancient
custom should be adopted of mixing in with them very swift infantry
with light shields, specially trained for the purpose, once called velites.
If this is done, no matter in what force the enemy cavalry turn out, they
cannot match the mixed formation.^ All ancient generals found this to
be the only answer. They trained young men who were outstanding

'V. says relatively little about cavalry or mixed formations of horse and light infantry,
because these tactics were highly developed and successful in his own day, cf 1.20 init.,
111.26 fin. He intends his legionary cavalry to include both light and heavy divisions, and
the latter at least seem to comprise two types, loricati and contati. Note the variety of
Arrian’s auxiliary cavalry, id. Ect. 21.

^loricati.

^contati.

^This word is identified as Celtic by philologists, cf. Kempf, 369-370, who explained it
as Celtic, cf. Old Irish “drong”, and this is preferred by Walde-Hofmann, 374, and
declared “sans doute” by Emout-Meillet, 185. The loose cavalry formation oy globus was
adopted by the Romans, too (cf 111.19), and the barbarian term bpouyyoc; became
standard for units or formations in the early Byzantine army, cf Maur. Strat. 1.3.10 etc.,
Leo Tact. 16.6. V. is the earliest author apart from the probably contemporary SHA Prob.
19.2 to use the word.

^globi, flying platoons.

^Thuc. 2.11.4: dSriXa yap id teSv TioX^ptov, Kai 6A.{yoo id 7roA-X.d icai 6i’ dpyfjc;
ai ^TrixEipfjcreiq yiyvovTai- TioXXdKK; xe to SJiaacrov 7iA.fiOoc; SESioq dpeivov fipovaxo
tout; TtXeovat; 5id to KaTa<j)povou" VTa<; driapaaKeu oo(; yeveaOai. “Events in war are
unforeseen, and attacks usually happen swiftly and in anger; often too the smaller force
with its fear fights better against the greater force because their arrogance makes them
unprepared.” Cf. Sail. Jug. 107.1: saepe antea <a> paucis strenuis advorsum
multitudinem bene pugnatum. “Often before now a few strong men have fought
successfully against a whole host.” The divinity V. alludes to is presumably fickle
Fortune, cf 111.22. V. is coy about the pagan religion, cf IV.39.

^mixtum agmen.

BOOK III: FIELD STRATEGY AND TACTICS

runners, placing them one between two horsemen, on foot and armed
with light shields, swords and javelins.’

17. On reserves,^ which are posted behind the line.^

The best principle, and that which contributes most towards victory, is
for the general to hold in readiness behind the line the pick of the
infantry and cavalry, together with unattached “vicars’\ “counts'” and
tribunes,'’ some about the wings, and some about the middle. Then
wherever the enemy attack strongly, they may rapidly move in to
prevent the line being broken, reinforce any weak points and, with their
additional strength, break the enemy’s onset.

This tactic was first discovered by the Spartans, imitated by the
Carthaginians, and later used everywhere by the Romans. No disposition
has been found better than it. For the straight battle-line ought to and
can only repel or rout the enemy.^ If a “wedge”^ or “pincer”^ is to be
formed, you will need to hold reserves behind the line, from which to

'There were contemporary kinds of mixed formation to which V is perhaps assimilating
the velites of the middle Republic, cf. Amm. 16,11 9 19.3.1, 21 12.9, 24,1.13: Caes. Bell.
Gall 1.48.5-7 describes a similar German mixed formation to that described by Amm.
16.12.21 at the battle of Strasbourg, AD 357 Caesar later used a similar tactic at the battle
of Pharsaius, 48 BC, cf. Front. Strat. 2.3.22. The velites are further described in action
against elephants at 111.24, as at the battle of Zama, 202 BC The original velites were
probably abolished by Marius’ reforms, cf. Festus 274.21 L.: parmulis pugnare velites
{milites, codd.) soliti sunt; quarum usum sustulit Marius datis in vicem earum Bruttianis,
'^Velites {milites, ‘soldiers’, codd.) were accustomed to fight with small round shields;
Marius phased out their use, giving instead Bruttian shields”, or at least soon after. Cf.
M.J.V. Bell, “Tactical Reform in the Roman Republican Army”, Historia xiv (1965) 421,
who argued from Front. Strat. 2.3.17, mentioning Sulla’s deployment of velites et levem
armaturam at Orchomenus in 85 BC, that Festus meant that Marius merely gave the velites
bigger shields.

^subsidia.

'Cf. Onas. 22.

^vicarii, comites tribunique vacantes, i.e. all staff-officers, cf III.IO. Note that Ammianus
applies vacans only to tribunes.

^Cf III praef on Spartan tactics mediated to the Carthaginians by Xanthippus and Sosylus
in the First and Second Punic Wars. Cato was said by Pliny Hist. Nat. praef 30 to have
learned his soldiering under Scipio Africanus and even Hannibal also. Each developed and
refined the offensive reserve over the Second Punic War, cf H.H. Scullard, Scipio
Africanus in the Second Punic War (Cambridge 1930) 212, 270, after Hannibal had used
it at Cannae, 216 bc, cf Livy 22,47.6-10.

^cuneus. Cf 1.26.

yorfex.

100

VEGETIUS’ EPITOME OF MILITARY SCIENCE

make your wedge or pincer. If a “saw”' is to be drawn up, it is likewise
drawn from reserves. Once you start transferring soldiers of the line
from their stations, you will throw everything into confusion. If a
detached “group”^ of the enemy begins to press your wing or some
other part, unless you have reserves which you can send against the
“group”, you will have to remove foot or horse from the line, and in so
doing you will denude one part at your peril in your desire to defend
another.^

When an abundant supply of soldiers is not available to you, it is
better to have a shorter line, provided you place a very large number in
reserve. In the middle of the field you need to hold the pick of your
heavy-armed infantry in reserve, with which to form a “wedge” and
suddenly break through the enemy line; whereas on the flanks you
should keep in reserve cavalry pikemen and cuirassiers, with the light¬
armed infantry, for surrounding the enemy wings.^

18. In what position the commander-in-chief, the second- and third-
in-command should stand.^

The general who holds the chief command usually stands between the
infantry and cavalry on the right flank. This is the position from which
the whole line is commanded, and from which there is direct and
unobstructed forward movement. He stands between the two arms so as
to be able to direct with his advice and exhort by his authority both
cavalry and infantry to battle. It is his task to use his cavalry reserves^

^serra.

^globus, flying platoon.

^These special tactics were described in Cato’s lost de Re Militari, cf. fr. 11= Fest.
466.30 L.: sive opus sit cuneo aut globo out forcipe aut turribus aut serra, uti adoriare,
“whether there is need of a ‘wedge’, ‘flying platoon’, ‘pincer’, ‘towers’ or ‘saw’ in order
to attack.” They are further discussed in III.18-20. Arr. Ect. 30 proposed to lengthen the
lines to deal with Alan flying platoons.

'‘This contradicts III. 16 ink., where only light cavalry are said to be for this purpose.

^This system of command may refer to the rotation-system followed for example at
Cannae. Cf. Livy 22.45.8.

^supernumerarii, probably merely one of V.’s non-technical word for reserves, drawn
from the usage of the civil service; cf 11.19 n. In III.17 he uses superjlui, abundantes,
reservati for the same, and in 111.18, superjlui and supernumerarii.

BOOK III: FIELD STRATEGY AND TACTICS

101

with light infantry mixed in with them to surround the enemy’s left
wing, which stands opposite himself, and press it constantly from the
rear.

The second-in-command is posted in the middle of the infantry line
to sustain and encourage it. He should have about him very strong and
heavy-armed infantry from the said reserves, from which to make a
“wedge” and break through the enemy line or, if the enemy form a
“wedge”, to make a “pincer” so he can counter their “wedge”.

On the left flank of the army the third-in-command should be found,
a suitably warlike and resourceful officer because the left flank is more
awkward and stands as though maimed^ in the line. He should have
about him good reserve cavalry and very swift light infantry, with which
he can constantly extend the left wing to prevent it being surrounded by
the enemy.

The war-cry, which they call barritus, should not be raised until both
lines have engaged each other. It is a mark of inexperienced or cowardly
men if they cry out from a distance. The enemy are more terrified if the
shock of the war-cry is made to coincide with the blows of weapons.^

Always strive to be first to draw up the line, because you can do at
your pleasure what you judge useful to yourself, while no one is
obstructing you. Secondly, you increase the confidence of your men and
diminish the courage of the enemy, because the side which does not
hesitate to challenge appears the stronger. The enemy, by contrast, begin
to be afraid when they see lines being drawn up against them. Thirdly,
it allows of the greatest advantage in that you may attack first while you
are drawn up and prepared and the enemy is still ordering his forces and

'Mancus, technically denoting one forced to use his left hand because of damage to his
right, cf. Ulp. Dig. 21.1.12.3. Cf. Thuc. 5.71 on the tendency of opposing battle-lines to
shunt respectively to the right.

^Originally the tribal war-cry of the Cornuti, either a tribe or a collective noun for
“Germans”, enrolled by Constantine I, the barritus quickly spread to the whole field army;
cf. Amm. 31.7.11, 26.7.17, 16.12.43, Lact. Plac. in Stat. Theb. 4.394 J., Kempf, art. cit.
372; A. Alfbldi, “Cornuti: A Teutonic contingent in the service of Constantine the Great”,
DOP xiii (1959) 169-179. Cf. also SHA Alex. Sev. 53.8-9, Arr. Ect. 25, Onas. 29; and
Ait. Tact. 44.1 on the various ethnic war-cries in use in the Roman army of Hadrian.

102

VEGETIUS’ EPITOME OF MILITARY SCIENCE

unsteady. For part of victory consists in throwing the enemy into
confusion before you fight.

19. Remedies to counter the strength and stratagems of the enemy
in battle.'

With the exception of raids and surprises, undertaken at some moment
of advantage which an experienced general never misses—for when
soldiers are marching wearily, divided while crossing a river, bogged
down in marshes, struggling over mountain-passes, dispersed and
unsuspecting in fields or sleeping in a night camp,^ an attack is always
made on favourable terms, as the enemy is preoccupied with other
business and is killed before he can prepare himself,^—[but] if the
opposition are careful and present no opportunity for ambush, then one
fights on equal terms, in the full presence, knowledge and view of the
enemy. However, the art of war is not less useful to trie proficient in the
open field of battle than it is in covert operations. Lake care above all
that your men be not surrounded on their left wing or flank, as often
occurs, or indeed on their right, although thi happens rarely , by a mass
of the enemy or by mobile ‘ groups'which they call If this

occurs there is one remedy: fold back and round off your wing or flank,
until your men who have wheeled round may defend the backs of their
comrades. At the salient angle of the end itself let a very strong force
be posted, because that is where the main attack is usually made,^

‘Further discussion of tactics originally described by Cato; cf III. 17. The use of the
modem tenn drungi shows V. is updating, as usual. Lang moved the rubric for III. 19
down to the sentence beginning Tamen, “however”, against the unanimous witness of the
MSS, and in violation of the best sense of V.’s words, which involve an anacoluthon after
the long parenthesis by way of stressing the point that a pitched battle ought to be a last
resort. Stelten restored the capitulation but incongruously kept Lang’s punctuation.

^mansio.

^V. hammers home the importance of opportunities for guerilla-tactics; cf. III.6, 111.8,
III. 10, 111.22, and Polyb. 9.12.7-8 on the third principle of generalship, the mastery of
timing.

*globi.

^flying platoons.

^The ^7tiKd|^7tio<; Td^ig, “angular line,” of Greek tactics, described by Ael.
Tad 45-46, Onas. 21.1-2.

BOOK III: FIELD STRATEGY AND TACTICS

103

Likewise there are appointed methods of countering a “wedge” of the
enemy. A “wedge” is the name for a mass of infantry who are attached
to the line, which moves forward, narrower in front and broader behind,
and breaks through the enemy lines, because a larger number of men are
discharging missiles into one position.‘ Soldiers call this tactic a “pig’s
head”.^ Against this is deployed the formation known as a “pincer”. A
body of crack troops is formed into a letter V, and this receives the
“wedge”, shutting it in on either side. Once this is done, it cannot break
through the line. The “saw” is the name of a formation which is ranged
by crack troops before the front, facing the enemy, so that a disordered
line may be repaired.^ The “group”"^ is a body of men who separate off
from their own line, and charge into the enemy in a mobile attack. A
more numerous and stronger globus is sent against it. Beware also of
deciding to change your ranks or transfer certain units from their
stations to others at the moment when battle is being joined. Uproar and
confusion instantly ensue, and the enemy more easily press upon
unready and disordered forces.

'See F. Lammert, “Der Keil in der Taktik des Altertums”, Gymnasium li (1940) 15-31,
identifying it as a double linear formation only vaguely resembling a wedge. Hannibal
used a cuneus at Cannae, cf. Livy 22.47.5 and 8.

^caput poreinum. Cf. Amm. 17.13.9; desinente in angustum fronte, quern habitum caput
porci simpUcitas militaris appellate “a front tailing off to a comer, which formation simple
soldiers call a ‘pig’s head’”, Agathias 2.8.8 (Keydell, 51): auoq Ke^Kx^n,
“pig’s head”, and Kempf, 369, citing Germans in Scandinavia who had a
svinjylkmg ~ acies porcina.

’Cf. Festus 466.28 L.: SERRA PROELIARI dicitur, cum assidue acceditur recediturque,
neque ullo consistitur tempore. “‘To fight in saw-formation’ is said when there is a
constant advancing and withdrawing, and no standing still at any time.”

^globus, flying platoon.

104

VEGETIUS’ EPITOME OF MILITARY SCIENCE

20. How many modes for engaging in a pitched battle there are, and
how the side that is inferior in numbers and strength may prevail.
There are seven types or modes of general actions,' when hostile
standards engage from two sides. The first action has the army in
rectangular formation with an extended front, just as even now it is
usual to do battle almost always.^ However, experts in military science
do not consider this type of action best, because when the line is
extended over a wide area, it does not always meet with even ground.
If there are any gaps in the middle, or a bend or curve, that is the point
at which the line is often breached.^ Further, if the enemy has the
advantage of numbers, he envelops your right or left wing from the
sides. There is great danger in this unless you have reserves who can
move up and hold back the enemy. Only he who has more numerous
and strong forces should engage in this formation. He should envelop
the enemy on both wings and enclose him as it were in the embrace of
his army.

The second action is oblique, and better in very many respects. With
this, if you draw up a small strong force in the proper position, you will
be able to bring off a victory even though you are impeded by the
numbers and strength of the foe. The method is as follows. When the
drawn lines advance to the encounter, you will remove your left wing
farther from the enemy’s right, so that neither missiles nor arrows can
reach it. You should fasten your right wing to the enemy’s left, and start
the battle there first, while using your best cavalry and most reliable
infantry to attack and surround the enemy’s left flank, on which you
have fastened, and by dislodging and outflanking them, reach the
enemy’s rear. If once you begin to rout the enemy from then on, with

^depugnationes.

^The contents of this chapter are probably largely based on Cato de Re h4ilUan\ cf. Jordan
fr. 10 = Non. Marc. 204.32 M.: una depugnatio est Jronte longo, quadrato exercitu, “A
first mode of general action is with an extended front, the army in rectangular formation”,
and Vegetius: una depugnatio est fronte longa quadro exercitu. The archaic usage (?)
whereby unus seems here to head a list secundus, tertius, etc.^ thus containing the idea
primus, is reproduced by Vegetius, but not the archaic masculine gender of frons, and
quadrants is replaced by the late-Latin quadrus.

^Cf. Polyb. 18.31.5-7 on the similar problem of finding suitable ground for the
Macedonian phalanx to operate.

BOOK III: FIELD STRATEGY AND TACTICS

105

those of your men who are attacking you will undoubtedly gain a
victory, while the part of your army which you moved away from the
enemy will remain undisturbed. The lines in this mode of battle are
joined in the shape of a letter A or a mason’s plummet-level.' If the
enemy does this to you first, you should assemble on your left wing
those cavalry and infantry we said should be placed behind the line as
reserves,^ and then resist the adversary with maximum force to avoid
defeat through tactics.^

The third action is similar to the second, but inferior insofar as you
begin by engaging his right from your left wing. For its attack is as it
were maimed/ the men fighting on the left wing are clearly in
difficulties when attacking the enemy. I shall explain this more clearly.
If you find that your left wing is far superior, reinforce it with your
strongest cavalry and infantry, and in the encounter attach it first to the
enemy right wing and make as much haste as you can to defeat and
surround the right flank of the enemy. But the other part of your army,
in which you know you have inferior warriors, remove as far as possible
from the enemy left, so that it is not attacked with swords or reached by
missiles. With this formation, care must be taken that your transverse
line shall not be harmed by wedge-formations of the enemy. It will only
be useful to fight in this mode in the case where your adversary has a
weak right wing, and you have a far stronger left.

The fourth action is as follows. When you have ordered your line, at
400 or 500 paces^ before you reach the enemy, suddenly spur on both
your wings when he is not expecting it, in order to turn the enemy to
flight by catching him unprepared on both wings and win a quick
victory. But although this type of battle may overcome quickly,
provided you deploy highly experienced and brave men, it is
nevertheless risky, because he who uses this formation is forced to

'i.e. the lines meet at an acute angle at one end; the cross-bar of the A is ignored.

'Cf. 111.18.

'Cf. Onas. 21.8.

'Cf. 111.18

^i.e. 2,000 or 2,500 Roman feet = c.647 yds. (591 m.) or c.808 yds. (739 m.), cf. III. 15
note.

106

VEGETIUS’ EPITOME OF MILITARY SCIENCE

denude the middle of his line and divide his army into two halves.
Moreover if the enemy is not beaten at the first assault, he has
opportunity to attack the divided wings and undefended middle line.'

The fifth action is like the fourth, but with the sole refinement of
placing the light armament and archers before the front line, so it cannot
be breached while they defend it. The general is then free to use his
right wing to attack the enemy left and his left wing to attack the enemy
right. If he can turn him to flight, he wins at once; if not, his middle
line does not come under pressure, being defended by the light
armament and archers.

The sixth action is the best, being very similar to the second. It is
used by those generals who despair of the numbers and bravery of their
men, and if they draw up their men well, they always win a victory
even with fewer forces. For when the drawn line nears the enemy, apply
your right wing to the enemy left, and start the battle there using very
reliable cavalry and very swift light infantry. Remove the remaining part
of your army as far as possible from the enemy line, and extend it in a
straight line like a spit. Once you begin cutting down the enemy left
wing from the flank and rear, you will certainly put them to flight. But
the enemy is prevented from assisting his men in trouble either from his
right flank or from the middle of his line, because your line is extended
and projects as a whole like a letter I, and recedes a very long distance
from the enemy. This formation is often used in encounters on marches.

The seventh action aids the combatant by using the terrain. This also
allows you to hold out against the enemy with fewer and less brave
forces. For example, if you have on one side of you a mountain, sea,
river, lake, city, marshes or broken country, so that the enemy cannot
approach from that direction, draw up the main part of your army in a
straight line, but on the side which does not have protection place all
your cavalry and light infantry. You may then safely engage the enemy

'Cf. Onas. 21.5-7. The tactic was used by the elder Scipio Africaniis against Hasdrubal
at the battle of Ilipa, 206 BC, cf Livy 28.14, Polyb. 11.20-24, cf Seneca de Vit. Beat.
4.1.

BOOK III: FIELD STRATEGY AND TACTICS

107

at your pleasure, because on one side the nature of the terrain protects
you, and on the other there is roughly double cavalry posted.'

But note this—nothing better has been found: if you intend to fight
with your right wing alone, place your strongest men there. If with your
left, station the most effective there. If you wish to form “wedges” in
the middle to breach the enemy lines, draw up your most experienced
soldiers in the “wedge”. Victory is usually owing to a small number of
men:^ so important is it for picked men to be ranged by a highly skilled
general in those positions which judgement and utility demand.

21. An escape-route should be offered to the enemy so that they
may be more easily destroyed in full flight.^

Most people ignorant of military matters believe the victory will be
more complete if they surround the enemy in a confined place or with
large numbers of soldiers, so they can find no way of escape. But
trapped men draw extra courage from desperation, and when there is no
hope, fear takes up arms. Men who know without a doubt that they are
going to die will gladly die in good company."

For this reason Scipio’s axiom has won praise, when he said that a
way should be built for the enemy to flee by.^ For when an escape-route
is revealed, the minds of all are united on turning their backs, and they
are slaughtered unavenged, like cattle.^ Nor is there any danger for the
pursuers once the defeated have turned round the arms with which they

'Cf. Onas. 21.3^.

^Cf. the iriarii, 1.20.

^This chapter has echoes of Sail. Cat. 58, Catiline's speech to his men when finally
surrounded by the senatorial forces in 63 BC.

'*Cf. Sen. Contr. 9.6.2: morientibus gratissimum est commori. “It is most gratifying to the
dying to die in good company,”

^Front. Strut. 4.7.16: Scipio Africanus dicere solitus est hosit non solum dandam esse
viam ad fugiendum, sed etiam muniendam, “Scipio Africanus used to say that a way
should not merely be given to the enemy to flee by, but even built for them”, cf ibid.
2.6.5,—the “golden bridge” of later tradition.

^’Sall. Cat. 58.21: cavete inulti animam amittatis, neu capti potius sicuti pecora
trucidemini... “Be warned not to lose your lives unavenged, nor be taken instead and
butchered like cattle.” Cf Livy 25.16.19.

108

VEGETIUS’ EPITOME OF MILITARY SCIENCE

could have defended themselves.' In this tactic, the greater the numbers,
the more easily is a mass cut down. For there is no need of numbers in
a case where the soldiers’ minds, once terrified, wish to avoid not just
the enemy’s weapons but his face.^ Whereas trapped men, though few
in number and weak in strength, for this very reason are a match for
their enemies, because desperate men know they can have no other
recourse. “The only hope of safety for the defeated is to expect no
safety.”^

22. How to retreat from the enemy if the plan to fight is rejected.
Now that I have summarized everything that military science preserved
from experience and theory, one matter remains for me to explain—how
to retreat from the enemy. For those learned in the art of war and
historical precedent affirm that nowhere is the threat of danger greater.
The general who retreats from the line before the encounter both
diminishes confidence among his own men and gives courage to the
enemy. But since this must often happen, the means of safely achieving
it need exposition.

First, your men should not know that you are retreating because you
refuse to enter battle, but believe they are being called back as part of
some strategy to attract the enemy onto more favourable ground and
overcome him more easily, or at least to set up a concealed ambush for
the enemy troops who follow. For men are ready to take flight if they
feel that their general is in despair. The enemy must also be prevented
from noticing your retreat and attacking at once. To this end many
generals have placed their cavalry in front of the infantry, manoeuvring
about so that the enemy might not see the infantry retiring. They then

'Sail. Cat. 58.16: nam in Juga salutem sperare, quom arma quibus corpora tegitur ab
hostibus avorteris., ea vero dementia est., “For to hope for safety by taking flight, when
you have turned away from the enemy the arms by which bodies are protected,—that truly
is madness.”

^Cf. Caes. Bell. Gall. 1.39.1: saepe numero sese cum his congresses ne vultum quidem
atque aciem oculorum dice bant ferre potuisse, “They would say that often in their
encounters with these men they could not bear even their face and the gaze of their eyes.”

Werg. Aen. 2,354; it is preserved only in the e group of MSS, Since V. likes to
quote Vergil, however, Lang was wrong to excise it.

BOOK III: FIELD STRATEGY AND TACTICS

109

removed and called back each line piecemeal, beginning with the
foremost, while the remainder stayed in their ranks. These were
gradually pulled back later to join those removed first.'

Some would retire with the army by night along routes they had
reconnoitred; when the enemy realized at dawn, they were unable to
overtake those who had gone ahead. Or else, the light armament were
sent ahead to the hills to which the whole army was withdrawn
suddenly, and if the enemy wished to pursue, they were routed by the
light armament who had occupied the place beforehand, aided by the
cavalry. For nothing is thought more perilous than for careless pursuers
to be attacked by men lying in ambush or who have prepared
themselves in advance.

This is a time when ambushes may be opportunely set, because over-
confidence and too little caution are used against fugitives. Necessarily,
more freedom from fear generally brings with it graver danger.^ Attacks
are usually made on those who are unready, whilst they are eating a
meal, wearily marching, grazing their horses and suspecting nothing of
the kind. This should be avoided by us and damage inflicted on the
enemy on such occasions. For neither bravery nor numbers can assist
those caught in such a case. He who is beaten in battle in a general
engagement, though there too art is of very great advantage, can
nevertheless in his defence accuse Fortune; he who suffers sudden
attack, ambushes or surprises cannot acquit himself of blame, because
he could have avoided these things and discovered them beforehand
through good scouts.^

On retreats the following stratagem is often used.'* A few cavalry
pursue by the direct route, while a strong force is sent secretly through
other localities. When the cavalry reach the enemy column, they
skirmish lightly with them and depart. Their general thinks he has
passed whatever ambush there had been and, freed from care, relaxes

'This is a stratagem of L. Aemilius Paullus, employed in 168 BC; cf. Livy 44.37.1-3.
'Cf. Thuc. 2.11.3-5.

^speculatorss, cf II1.6.

^i.e. against the retreating party.

VEGETIUS’ EPITOME OF MILITARY SCIENCE

into incaution. Then the force sent by the secret route attacks and
destroys them unexpectedly. Many generals retreating from the enemy,
if they are to go through forests, send men ahead to occupy defiles and
precipitous places, to avoid falling into ambush there. They also block
roads behind them by felling trees, in what they call a concaedes.^
depriving the enemy of the facility to pursue.^

Opportunities for ambush on the march are pretty well equally shared
between both sides, for the (adversary) in front leaves traps behind him
in suitable valleys and wooded mountains, and when the enemy falls
into them returns himself to assist his own men. The pursuer for his part
sends light troops far ahead by back paths and prevents the adversary up
ahead from getting past, closing him in, trapped in front and behind. If
the opposition sleep at night either the (adversary) in front can double
back, or the pursuer can attack by surprise, although there is a distance
between them. At the crossing of rivers let the party in front try to rout
the division that crosses first, while the rest are still separated by the
river-bed, or let the pursuer arrive by forced marches to discomfit those
who have not yet been able to cross.^

23. On the camels and armoured cavalry."^

Some nations in ancient times led forth camels into battle, and the
Urcilliani in Africa and the other Mazices lead them forth even today.^
It is a type of animal well-adapted to sands and enduring thirst, and is
said to keep straight on roads without error even when they are

‘“abattis”.

^Cf Tac. Ann. 1.50, Amm. 16.11.8, 17.1.9, 17.10.6.

^Cf Caes. Bell. Civ. 3.75.3-5, Maur. Sirat. 4 HEPl ’ENEAPAI. “On Ambush.”

^Antiochus III, king of Syria, appears to have been first to field cataphract cavalry and
camels against the Romans at the battle of Magnesia, 190 BC, cf Livy 37.40.5, 11, 12,
37.42.2: alii propter gravitatem tegumentorum armorumque oppressi sunt, “others were
overcome because of the weight of their armour and weapons”, 37.40.12; cameli, quos
appellant dromadas, his insidebant Arabes... “camels called dromadaries, on these sat
Arabs...”

^The Urcilliani are otherwise attested only by Corippus lohann. 5.390, in the sixth
century ad.

BOOK III: FIELD STRATEGY AND TACTICS

111

obscured by dust in the wind. However, apart from its novelty when it
is seen by those not used to it, it is ineffective in battle.’

Armoured cavalry^ are safe from being wounded on account of the
armour they wear, but because they are hampered by the weight of their
arms are easily taken prisoner and often vulnerable to lassos.^ They are
better in battle against loose-order infantry than against cavalry, but
posted in front of legionaries or mixed with legionaries they often break
the enemy line when it comes to comminus, that is, hand-to-hand,
fighting.'*

*It is claimed that the Romans later made some use of camels in battle as dromadarii, as
well as for transport; cf Ps.-Hyg. 29, R.W. Davies, “Ratio and Opinio in Roman military
documents”, Historia xvi (1967) 117, M.P. Speidel, Guards of the Roman armies,
Antiquitas Reihe I, Bd, XXVIII (Bonn i978) 26; as used for transporting heavy or bulky
supplies, cf. Davies, BASF v (1968) 33-34. However, the arch of Galerius reliefs show
only transportation, cf. H.P. Laubscher, Der Reliefschmuck des Galeriusbogens in
Thessaloniki, Deutsches Arch. Inst. Archaologische Forschungen I (Berlin 1975) Taf 28.3
camels carrying booty (?), Taf. 14.1 dromadaries carrying captive Persian royal harem-
ladies.

^cataphracti equites.

^laquei. A Hun tactic, according to Amm. 31.2.9. The Parthians also had mounted
soldiers who fought using a lasso, cf Suda S,278 s.v. aeipafq, and were themselves
famous for their cataphract cavalry, cf. Suda © .439 s.v. 0c6pa^. Cf. IV.23 where it means
“nooses”.

^M.P. Speidel, “Horsemen in the Pannonian Alae”, 57xliii (1987) 61-65, = Roman Army
Studies II 62^6, id., “Catafractarii clibanarii and the rise of the later Roman mailed
cavalry: a gravestone from Claudiopolis in Bithynia”, EA iv (1984) 151-156, with further
refs., = Roman Army Studies II 406-413, J.R. Rea, “A Cavalryman’s Career, A.D. 382
(?>-40r’, ZPE Ixvi (1984) 80.

112

VEGETIUS’ EPITOME OF MILITARY SCIENCE

24. How scythed chariots and elephants may be resisted in battle.'
Scythed chariots^ were used in battle by king Antiochus^ and
Mithridates/ Although at first they caused much alarm, they soon
became a laughing-stock. For it is difficult for a scythed chariot to find
a field that is constantly flat, and it is hindered by a slight impediment
and captured if a single horse is stricken or wounded.*^ But most of all
they fell victim to the following tactic of the Roman soldiers. When
they came to battle, the Romans suddenly threw caltrops^ over the
whole field. The speeding chariots were destroyed as they encountered
them.^ A caltrop is a defensive weapon made out of four spikes;
whichever way you throw it, it stands on three spikes and is armed by
the fourth which is erect.

Elephants in battle cause men and horses to panic because of the size
of their bodies, the horror of their trumpeting and the novelty of their
very form. King Pyrrhus in Lucania was the first to use them against the
Roman army,* and later Hannibal in Africa,^ king Antiochus in the

'V. summarizes a traditional chapter. Arrian Tact. 2.5, cf. 19.1, found in his sources
along with elephants various kinds of scythed chariot, with armoured and unarmoured
horses, with a single pole and with two or more poles, but did not think the material
worth transmitting because of its obsoleteness. Servius adAen. 1.476, writing in the late-
fourth century, knew of scythed chariots only from ancient history. SHA Alex. Sev. 55.2,
56.4, on the capture by Severus Alexander during the Persian campaign of AD 232 of
1,800 scythed chariots, is apocryphal.

^quadrigae falcatae, according to tradition invented by Cyrus the great, king of Persia,
Xen. Cyr. 6.1.29, cf. Arr. Tact. 22, Anon, de Rebus Bellicis 12.1.

^Antiochus ill, king of Syria, used scythed chariots at the battle of Magnesia, 190 BC, cf
Livy 37.41.5, Florus 1.24.16.

'‘Mithridates VI Eupator, king of Pontus, whose general Archelaus used scythed chariots
at the battle of Chaeronea, 86 BC, cf Front. Strat. 2.3.17, Plut. Sulla 15.1, 16.3, and
Mithridates himself used them in northern Asia Minor against Lucullus in 73 BC, cf Plut.
Lucull 7.5. His son Phamaces used them against Julius Caesar at Zela in 47 BC, cf A.
Hirtius Bell. Alex. 75.2, the last reliable use of this weapon.

*The Anon, de Rebus Bellicis 12-14 attempted to solve these problems by means of
hinged blades and armoured horses, in his highly fanciful military reforms proposed to a
late-fourth- or early-fifth-century ad Emperor.

^'tribuli.

^The reference appears unknown. On tribuli cf. Philo Byz. Bel. 100.7, Polyaen. 1.39.2,
4.3.17, Herodian 4.15.2-3.

’’At the battle of the river Siris, at Heraclea, Lucania, 280 BC, cf Plut. Pyrrh. 16-17,
Dion. Hal. 19.4-12, and Florus 1.13.7-8 who wrongly says Heraclea in Campania and the
river Liris.

“^At the battle of Zama, 202 BC, cf Livy 30.33.1-3, 14-16.

BOOK III: FIELD STRATEGY AND TACTICS

113

Orient/ and Jugurtha in Numidia^ had them in large numbers. Various
methods of resistance have been worked out against them.—

A centurion in Lucania cut the hand off one with his sword,—what
they call the proboscis? Pairs of cataphract horses were harnessed each
to a chariot; mounted on (the horses) were cataphract cavalrymen"* who
aimed sarisae, that is, very long pikes,^ at the elephants. Being covered
in iron they were not harmed by the archers riding on the beasts, and
avoided their charges thanks to the speed of their horses. Others sent
against elephants cataphract infantrymen;^ on their arms, helmets and
shoulders huge iron spikes were set, so that the elephant could not use
its trunk to catch hold of the soldier coming against him. But especially
the ancients deployed velites^ against elephants. Velites were young men
lightly armed and able-bodied, who sent spears with marvellous skill
from horseback. While the horses ran past, they brought the beasts down
using broad lances* or large javelins;^ then, with increasing boldness,
larger numbers of soldiers would combine together to cast pila, that is,
javelins, into the elephants, destroying them with wounds. Another
method was for slingers with “sling-staves”and slings to shoot round
stones at the Indians controlling the elephants, knock them off, turrets

'At the battle of Magnesia, 190 or 189 BC, cf. Livy 37.40.2^, Florus 1.24.16.

^At the battle of the river Muthul, 109 BC, cf. Salt. Jug 49.1, 53.3^.

M.e. trunk. The word is misspelled promuscis in the MSS, as also in the MSS of Florus
1.13.9. The allusion is to an incident at the battle of Asculum, in Apulia not Lucania, 279
BC, cf. Florus ibid.

^clibanarii. V. uses a late-Roman (from Persian) term for them; cf. RE IV (1900) 21-22
s.v. (Fiebiger), M.P. Speidel, “Catafractarii clibanarii and the rise of the later Roman
mailed cavalry: a gravestone from Claudiopolis in Bithynia”, EA iv (1984) 151-156 = id.,
Roman Army Studies II 406-413.

^conti. Presumably the chariot was used to screen the backs of the attackers, cf. Anon, de
Rebus Bellicis 12-14. On the saris(s)a, see M.M. Markle, “The Macedonian Sarissa,
Spear, and related armor”, AJA Ixxxi (1977) 323-339; P.A. Manti, “The Macedonian
Sarissa, Again”, AncfV xxv (1994) 77-91.

^'cataphracti milites.

^light troops.

^lanceae.

'^spicula. The size of the weapons is explained below by that of the ballista-bolts as
matching the elephants. The role of velites is also described at 111.16, where they perform
on foot; the material obviously belongs to the middle Republic.

^^^Jiistibali, cf. III. 14.

114

VEGETIUS’ EPITOME OF MILITARY SCIENCE

and all, and slay them; no safer method has been found than thisJ Or
else, as the beasts charged, the soldiers yielded ground to them as if
they had broken into the line. When they reached the midst of the
formation, they were surrounded on all sides by massed groups^ of
soldiers and captured with their drivers, intact and free from wounds.^
It is advisable to post behind the line carriage-ballistas'* of a somewhat
larger model—^these shoot bolts farther and with greater force—mounted
on cars with pairs of horses or mules; then, when the beasts come
within the weapon’s range they are pierced by ballista-bolts.^ A broader
and stronger iron head is fitted so as to make larger wounds in large
bodies. Against elephants we have listed several examples and devices,
so that if the need ever arise it may be known what should be deployed
against such monstrous beasts.^

25. What to do, if part or if all of the army is routed.

Note that if part of the army is victorious and part is routed, one should
be hopeful, because in a crisis of this type the steadfastness of the
general can reclaim the whole victory for himself. This has happened in
countless battles, and those who have despaired the least have been
taken for the winners. Where their situations are similar, he is judged
the stronger who is not dismayed by his adversities. Let him be first,
therefore, to take spoils from the enemy slain—as (the soldiers)
themselves say, “collect the field”.^ Let him be seen to be first to cele¬
brate with shouts and bugles. By this show of confidence he will terrify

'The mahouts were called “Indians” by the Romans regardless of whether they controlled
Indian or African elephants. Cf. Polyb. 1.40.15, Livy 38.14.2. See also Apoc. 1 Macc.
6.38.

^globiy massed platoons.

’The stratagem was Scipio’s at Zama, 202 BC, cf Livy 30.33.1-3, 14-16, Polyb.
15.9.6-10, 15.12.1^, Front. Strat 2.3.16.

*carroballistaey cf. 111.14 for carroballistae behind the lines.

^sagittae ballistariae.

‘^Arr. Tact 19 dismissed chariots and elephants as obsolete, diverging from Ael. Tact 22.
But the Persians fielded elephants (but not chariots) regularly in the fourth century AD,
cf Amm. 19.2.3, etc., who also finds them formidable.

^colligat campum. This piece of soldiers’ slang is not otherwise attested; cf Kempf, 376.

BOOK III: FIELD STRATEGY AND TACTICS

115

the enemy and double the confidence of his own men, as if he had come
off victor in every part of the field.

But if for some reason the whole army be routed in battle, the
disaster can be mortal; yet the chance to recover has existed for many,
and a remedy should be sought. Let the provident general therefore
engage in pitched battle only if he has taken precaution that, should
something go wrong owing to the variability of wars and the human
condition, he may still get the defeated away without great loss. For if
there are hills nearby, if there are fortifications to the rear,' or if the
bravest troops resist while the rest retreat, they will save themselves and
their comrades. Often a previously routed army has recovered its
strength and destroyed those in loose order and pursuing at random.
Never does greater danger tend to arise for the side that is celebrating,
than when over-confidence is suddenly turned to panic.^

But in any event survivors should be collected up, stiffened for war
by means of appropriate exhortations and restored with new arms. Then
new levies of legionaries^ and new auxilid^ should be sought and, what
is more important, opportunities to attack the victors themselves through
concealed ambushes should be exploited, and morale regained in this
way. There is no lack of opportunity, since human minds are carried
away by success to become so much the more arrogant and careless. If
anyone shall think this his last chance, let him reflect that the results of
all battles in the early stages (of a war) have been more against those
destined to ultimate victory.^

■Cf. 1.21.

'Cf. Onas. 36.3-6, Thuc. 2.11.4.
^dilectiis.

^palatine auxilia.

^Cf. Polyb. 3.4.4-5.

116

VEGETIUS’ EPITOME OF MILITARY SCIENCE

26. General rules of war.'

In all battles the terms of campaign are such that what benefits you
harms the enemy, and what helps him always hinders you. Therefore we
ought never to do or omit to do anything at his pleasure, but carry out
only that which we judge useful to ourselves. For you begin to be
against yourself if you copy what he has done in his own interest, and
likewise whatever you attempt for your side will be against him if he
chooses to imitate it.

In war, he who spends more time watching in outposts and puts more
effort into training soldiers, will be less subject to danger.'

A soldier should never be led into battle unless you have made trial of
him first.^

It is preferable to subdue an enemy by famine, raids and terror, than in
battle where fortune tends to have more influence than bravery
No plans are better than those you carry out in advance without the
enemy’s knowledge.^

Opportunity in war is usually of greater value than bravery.^

In solliciting and taking in enemy soldiers, if they come in good faith,
there is great security, because deserters harm the enemy more than
casualties.’

It is preferable to keep additional reserves behind the line than to spread
the soldiers too widely.®

It is difficult to beat someone who can form a true estimate of his own
and the enemy’s forces.^

‘V.’s “general rules” were much the most popular part of the Epitome in the Middle
Ages, and Maurice at the end of the sixth century translated a number of them into Greek
in his Strategica 8.2, as well as adding many more, some taken from elsewhere in V.’s
Epitome, others from Onasander and other sources.

'Cf. 1.3.

'Cf. III.9, III.10.

^Cf. I1I.3, III.9, III.22.

'Cf. I1I.6, III. 18.

"Cf. III. 18 fm.

"Cf. 111.6 fin.

’’Cf 111.15.

‘'Cf. 111.9.

BOOK III: FIELD STRATEGY AND TACTICS

117

Bravery is of more value than numbers.’

Terrain is often of more value than bravery.^

Few men are bom naturally brave; hard work and good training makes
many so.^

An army is improved by work, enfeebled by inactivity.'*

Never lead forth a soldier to a general engagement except when you see
that he expects victory.^

Surprises alarm the enemy, familiarity breeds contempt.^

He who pursues rashly with his forces in loose order is willing to give
the adversary the victory he had himself obtained.^

He who does not prepare grain-supplies and provisions is conquered
without a blow.^

He who has the advantage in numbers and bravery, let him do battle
with a rectangular front, which is the first mode.^^

He who judges himself unequal, let him rout the left wing of the enemy
with his right, which is the second mode.’^

He who knows he has a very strong left wing, let him attack the right
wing of the enemy, which is the third mode.”

He who has very experienced soldiers should begin battle on both wings
together, which is the fourth mode.'“

He who commands an excellent light armament, let him attack both
wings of the enemy after posting the light troops before the line, which
is the fifth mode.’^

'Cf. (.8, 111.16, 111.20 fm.
^Cf. 111.13.

’Cf. I.l, 1.2, and passim.
'Cf. 11.23 fin., 111.4.

'Cf. 111.12.

'Cf. 1II.6, 111.12.

'Cf. 111.22, m.25.

'‘Cf. rn.3, 111.9.

'Cf 111.20.

"’Cf 111.20.

"Cf 111.20.

'^Cf 111.20.

"Cf 111.20.

118

VEGETIUS’ EPITOME OF MILITARY SCIENCE

He who has confidence neither in the numbers of his soldiers nor their
bravery and is to fight a pitched battle, let him repel the left wing of the
enemy with his right having extended the rest of his men in the form of
a spit; which is the sixth mode.'

He who knows he has fewer and inferior forces, in the seventh mode let
him have on one flank a mountain, city, sea, or river, or some (other)
support.^

He who has confidence in his cavalry should find places more suited to
horsemen and wage war more by means of cavalry.^

He who has confidence in the infantry forces should find places more
suited to infantry and wage war more through infantry."

When an enemy spy^ is wandering secretly in camp, let all personnel be
ordered to their tents in daylight, and the spy is immediately caught.^
When you discover that your plan has been betrayed to the enemy, you
are advised to change your dispositions.^

Discuss with many what you should do, but what you are going to do
discuss with as few and as trustworthy as possible, or rather with
yourself alone.^

Soldiers are corrected by fear and punishment in camp, on campaign
hope and rewards make them behave better.^

Good generals never engage in a general engagement except on some
advantageous occasion, or under great necessity."^

It is a powerful disposition to press the enemy more with famine than
with the sword."

'Cf. III.20.

'Cf. m.9. in.13.

'Cf. in.9, III.13.

^explorator.

^’Not mentioned elsewhere in the text. Maur. Strut. 9.5.20 records the same tactic.
^Not mentioned elsewhere in the text.

'^Cf. III.6, III.9.

'Cf III.4.

'*’Cf III.9.

"Cf II1.3, III.9.

BOOK III: FIELD STRATEGY AND TACTICS

119

The mode in which you are going to give battle should not become
known to the enemy, lest they make moves to resist with any counter¬
measures.^

On cavalry there are many precepts, but since this branch of the
military has progressed in its training practices, type of armour^ and
breed of horses, 1 do not think there is anything to be gained from
books, for the present state of knowledge is sufficient.

I have set out, Invincible Emperor, the principles which the noblest
authors handed down to posterity as having won the approval of
different ages in the test of experience. To your skill at archery which
the Persian admires in Your Serenity, to the skill and grace of your
horsemanship which the nation of the Huns and Alans would like to
imitate if it could,^ to your speed in the charge which the Saracen and
Indian cannot match, to your expertise in armaturd^ even part of whose
routines the drillmasters^ are delighted to have appreciated (as
connoisseurs),^—may now be joined a Rule-book of Battle, or rather an
Art of Victory, in order that by the valour coupled with the strategy of
your glorious State, you may manifest your role of both Commander-in-
chief (Emperor) and Soldier.^

'Cf. III. 19, III.20.

^Cf. 1.20 init. on the improvements due to the Goths, Huns and Alans, and cf.
Mulomedicina 3.6.5, where the Hunnic breed of warhorse is given pride of place, and a
full description.

^The junction of these two peoples is reported by Amm. 31.3.1, cf Ps.-Victor Epit de
Caess. 47.3.

'special drill, cf 1.13 note, 11.23.

^campidoctores.

^The presence of these barbarians in the context of feats of arms including armatura, and
the emulation ascribed to them in the same breath as to the Roman drillmasters, may
suggest tournaments, i.e., friendly competition. In any event, the particular combination
of names suits best the reign of Theodosius I, who concluded peace-treaties with the king
of Persia, and with the Saracens and people called “Indians”, cf Pac. Pan. Lat. 2 (12)
22.2-5, as well as with the Huns and Alans, cf. ibid. 32.4, in the ad 380s.

’Cf. Sail. Cat. 60.4: strenui militis et boni imperatons officio simvl exequebatur, “he
fulfilled at once the duties of an energetic soldier and a good commander-in-chief”

Book IV

Preface.

The wild and uncivilized life of man at the beginning of time was first
separated from communion with dumb animals and beasts by the
founding of cities. In them the common utility discovered the idea of
the State. And so the most powerful nations and consecrated Princes
thought that there was no greater glory than the foundation of new cities
or the transfer of those founded by others to their own name under some
expansion.' In this enterprise the Clemency of Your Serenity wins the
palm. For by others few or single cities, by Your Piety countless cities
have been completed with such continuous activity that they appear not
so much built by human hands as sprung up by divine will.^ You
surpass all Emperors in felicity, moderation, morality, displays of
indulgence and love of studies.^ We see the good things of your reign
and mind and we hold what both the past age wished to anticipate and
the future longs to be perpetuated for ever."* We rejoice that in these
things as much has been granted to the whole world as human mind
could ask or divine favour bestow. But the value added to the
dispositions of Your Clemency by the elaborate construction of walls is
demonstrated by Rome, who saved the lives of her citizens through the
defences of the Capitoline citadel, to the end that she might later win a
more glorious Empire of the whole world.^ To complete, then, a work
undertaken by command of Your Majesty, I shall summarize in order
from various authors the measures by which either our cities should be

'i.e. of the cities.

^Symm. II laud, in Val. 20 uses civitates of towers. Julian Ep. ad Athen. 291A claimed
to have rebuilt 45 “cities”, probably including forts on the German frontier as well as
Gallic towns, cf Amm, 16.1 l.l I, 17.9.1; Themistius’ description of Valens building the
fort of Charsovo in Thrace 368-9, Or. io.l36a-138b: forts “rose as if by magic”. But
Theodosius 1 seems to have founded or renamed a significant number of cities in the East,
e.g. Theodosiopolis-Resaina in Osrhoene, Theodosiopolis in Armenia, Theodosiana in
Arcadia, Aegyptiaca, perhaps two other Theodosiopoleis in Egypt, Arcadiopolis-Bergula
in Thrace, etc. Other examples of “Theodosiopolis” or “Theodosiana” may have been
founded by Theodosius II. The historical exegesis seems based partly on Sail. Cat. 1-2;
cf also Hor. Serm. 1.3.99-112.

^Cf. Ps.-Victor Epit. de Caess. 48.11-12 on Theodosius I.

'‘Cf Aus. Grat. actio 16 (75); bonitatis et virtutis exempla, quae sequi cupiat ventura
posteritas et, si rerum natura pateretur, adscribi sibi voluisset antiquitas, “examples of
goodness and virtue, such as future posterity will desire to imitate and, if the nature of
things would allow, the past would have wished to be ascribed to itself.”

^One of three references to the siege of the Capitol by the Gauls in 390 bc (traditional
date), cf IV.9 and IV.26. Rome is a strong symbol of inviolability for V., which strongly
suggests he was writing before it was sacked by Alaric the Goth in ad 410.

BOOK IV: SIEGE AND NAVAL WARFARE

121

undertaken by command of Your Majesty, I shall summarize in order
from various authors the measures by which either our cities should be
defended or the enemy’s destroyed. Nor shall 1 regret my labour since
foundations are being laid for the benefit of all.

(Fortifications, ch. 1-6)

1. Cities should be fortified either by nature or by works.

Cities and forts are either fortified by nature or by human hand, or by
both, which is considered stronger. By “nature” is meant places which
are elevated, precipitous, surrounded by sea, morasses or rivers; by
“hand”, fosses and a wall. In places enjoying the safest natural
advantages, judgement is required of the selector; in a flat place, effort
of the founder. We see cities of very great antiquity so built in the midst
of spreading plains that in the absence of help from the terrain they
were rendered invincible by art and labour.'

2. Walls should be built not straight, but with angles.

The ancients refused to build the circuit of their walls in straight lines,
fearing that it might be exposed to the blows of rams. Instead they
enclosed cities within foundations laid out in sinuous windings,^ and put
more frequent towers precisely at the angles so that if anyone tried to
move up ladders or machines to a wall constructed on this system, he
would be shut in before, on the flanks and virtually behind as though in
an embrace, and destroyed,^

'e.g. Zama, Numidia, cf. Sail. Jug. 57.1: id oppidum, in campo situm, magis opere quam
natura munitam erat, nullius idoneae rei egens... “that settlement, situated in the plain,
had been fortified more by labour than by nature, was short of no needful thing...”

^Cf. Vitr. 1.5.2: Curandumque maxime videtur ut non facilis aditus sit ad oppugnandum
murum, sed ita circumdandum ad loca praecipitia et excogitandum uti portarum itinera
non Sint directa sed scaeva, “One should take particular care that access to the wall for
an attack should not be easy, but (the wall) should be so disposed around steep places and
so designed that the roads to the gates should be not straight but bending to the left”; cf
Philo Byz. Polior. 86.2-11, 79.7-16; Amm. 20.7.17, describing a fort in Mesopotamia
(Virta) thought to have been built by Alexander the Great; muris velut sinuosis
circumdatum et cornutis instructioneque varia inaccessum, “surrounded by walls almost
sinuous and cresent-shaped and made inaccessible by various construction.”

^Cf. Vitr. 1.5.2: Item turres sunt proiciendae in exteriorem partem uti, cum ad murum
hostis impetu velit adpropinquare a turribus dextra ac sinistra lateribus apertis telis
vulnerentur, “Likewise towers are to project forward into the exterior side so that when
the enemy wishes to approach the wall for an attack, they may be wounded by missiles
from the towers, which have openings in their right and left flanks”; cf Philo Byz. Polior.
79.11-16, 82.43-50.

122

VEGETIUS’ EPITOME OF MILITARY SCIENCE

3. How a terreplein* is attached to the wall from ground-level.

The wall is completed so that it can never be knocked down, on the
following plan. Two internal walls are built at intervals of twenty feet
each. Then earth which has been dug out of the fosses is dumped
between them and rammed solid with piles, the first internal wall from
the outer wall being built proportionately lower, the second far smaller,
so that one can ascend from the ground-level of the city as if by stairs
on a gentle slope up to the battlements. For no rams can breach a wall
that is strengthened by earth, and if for some reason the masonry should
be demolished the mass that has been compacted between the internal
walls resists intruders in the wall’s stead.^

4. On portcullises and gates, protecting them from fire.

Take care also that gates may not be burned by fires placed against
them. To counter this they should be covered with hides and iron. But
it is more useful, as antiquity discovered, to add a barbican^ before the
gate, and place in its entrance a portcullis,"^ which hangs on iron chains
and ropes. Then if the enemy enter, it is let down, and they are shut in
and exterminated.^ The wall above the gate should be so designed as to
admit openings through which water may be poured from the upper
storey to quench a fire below.

5. On making fosses.

Fosses before cities should be made as wide and as deep as possible,^
so that they cannot easily be levelled and filled in by besiegers and,
once they begin to overflow with floodwater, hardly permit a sap to be
continued by the enemy. A subterranean work is prevented from being
constructed in two ways, by the depth and by the flooding of fosses.

^adgestus, talus or terreplein; more commonly means a siege-mound.

’Cf. Vitr. 1.5.5-7, on which account V. does not draw.

^propugnaculum, fortification, bastion.

^cataractQy Greek for cataract, waterfall, portcullis.

^On portcullises, cf Aen. Tact. 39.3-4, Livy 27.28.10-12, describing the capture and
massacre of 600 Roman deserters at Salapia, 208 BC, Appian Bell Civ. 4.78 (328) on
Xanthus besieged by Brutus in 42 BC, Amm. 30.5.17 on Valentinian fs inability to leave
by a gate blocked probably by a jammed iron portcullis at Savaria in ad 375.

^’Cf. Vitr. 1.5.6, Philo Byz. Polior. 85.4-6, 91.19-24; neither mentions moats.

BOOK IV: SIEGE AND NAVAL WARFARE

123

6. On preventing men on the walls from being injured by enemy
arrows.

There is a risk of defenders being frightened off the fortifications by a
mass of archers, who then set up ladders and occupy the wall. Against
this as many as possible in cities should have cataphracts and shields.'
Secondly double-thickness cloaks and goat’s hair mats^ are strung up
along the battlements to absorb the impact of arrows. For darts do not
easily pass through material which yields and swings.^ Another remedy
was invented where they made frames of wood which they called
metallae,^ filled them with stones, and set them up between pairs of
battlements in such a way that if an enemy climbed up a ladder and
touched any part of one, he tipped the rocks onto his own head.^

(Preparations for a siege, ch. 7-11)

7. Measures to be taken so that the besieged do not suffer famine.
There are many forms of defence and attack, which we shall introduce
at the appropriate place. For the present, note that there are two kinds
of siege warfare; in one of them the adversary deploys pickets in
suitable positions and^ either prevents those under blockade from getting
water, or hopes for a surrender through famine once he has stopped all
transportation of supplies. By this strategy he himself remains at leisure
and safe, while he wears down the enemy.

'Cf. 1.20.

^Cilicia, literally “Cilician (fabrics)”, cf. IV. 14.

'V.’s explanation resembles Heron Mech. 2.34q (Nix-Schmidt 11.1, Teubner edn. 1900
p.l86). Cf, Amm, 24.2.10: per propugnacula ciliciis undique laxius pansis, quae telorum
impetus cohiberent, “goat’s hair mats were spread loosely everywhere along the
battlements, which might stop the impact of missiles”, cf. Livy 38.7.10, Sisenna Hist. fr.
107, Cf, also Aen. Tact. 32.1, 9-10 sails and thick curtains, Philo Byz. Potior. 95.34 skins
or curtains, Dexippus Scyth. fr. 29.3, F. Jacoby (ed.), FGH IIA p.474 cloaks and fleeces,
Jos. bell. lud. 3.172-173 raw hides, Thuc. 2.75.5 skins and hides.

'‘Perhaps matella, chamber-pot. Metalla is unattested, cf. TLL s.v

^Cf. Anon. Byz. Parang. Polior. 205.3, wicker baskets tilled with rubble or compacted
earth, but thrown from walls. Cf. also Apoll. Dam. Potior. 139.11-12.

^’A lacuna is noted here in the editions and supplemented in MS Parisinus 7232
(fourteenth century), coniinuis insuttibus impugnat obsess os, at ter am cum, “attacks the
besieged by means of unremitting assaults, the other (is) when (he)...” As V. does not
expressly talk about the other kind of siege warfare, the violent assault, until IV. 12 ff., It
is possible that there is no lacuna here, especially when IV.7-11 are all about the problem
of the interruption of supplies.

124

VEGETIUS’ EPITOME OF MILITARY SCIENCE

Against these calamities landowners, prompted by the least suspicion,
should as thoroughly as possible collect all means of sustaining life
within walls so that they may have an abundance of supplies for
themselves, and shortages force the adversaries to withdraw. Not only
pork, but every kind of animal which cannot be kept enclosed should be
sent for curing,‘ so as to eke out the grain with the aid of meat.
Farmyard fowls, however, are fed without cost in cities and are needed
for the sick. Fodder for horses above all should be stockpiled, and what
cannot be carried in should be burned; stocks of wine, wine-vinegar, and
other crops and fruits should be collected, and nothing which may be of
use left for the enemy. Considerations of both utility and morale urge
that gardening (for food) be conducted in the pleasure-grounds of town-
houses and (public) open spaces.^ But there is little point in having
collected large stocks without sensible issue from the outset by
competent officials: men who began to keep a frugal diet while there
was still plenty have never been in danger of starving.^ Also those unfit
for fighting by reason of age or sex were often shut out of the gates
because of the need to conserve food, lest hunger oppressed the soldiers
guarding the walls."*

8. What supplies should be procured for the defence of walls/

It is advised to procure bitumen, sulphur, liquid pitch and the oil which
they call “burning-oil”,^ for burning the machines of the enemy.^ For

Uaridum^ “bacon”. Cf. Macr. Sat. 7.12.2: saepe adposita salita came, quant laridum
vocamus, ut opinor quasi large aridam, quaerere mecum ipse constitui qua ratione carnem
ad diuturnitatem usus admixtio salis servet. “Often when salt meat has been served which
we call Marida’ (bacon) as ‘largely arid’, I suppose, I have resolved to ask myself by what
principle the admixture of salt preserves meat for prolonged use.”

^Cf. the siege of Cremna, AD 278, where houses were razed and the sites sown with grain,
Zos. 1.79.2.

’V. largely repeats material from 111.3. Since in the epilogue at IV.30 he draws attention
to his pronouncements on the subject of supplies, he may well have added his own ideas.
Certainly the background is contemporary, cf. Amm. 18.7.3, and the details are
circumstantial. But the basic strategy was that of Q. Fabius Maximus (Cunctator), dictator,
217 BC, cf. Livy 22.11.4.

'’Cf. Thuc. 2.78.3, Onas. 42.23, Front. Strat. 3.4.5.

^Cf. a similar chapter by Philo Byz. Polior. 89.46-90.24.

^'oleum incendiarium, cf Amm. 23.6.37 on the military use of “Medic oil” in fire-darts
(cf IV. 18), and 20.11.15 on the Persians pouring boiling pitch onto the Roman ram at
Bezabde, ad 360.

'Cf Jos. Bell. lud. 3.228, Amm. 20.11.18-19, 22-23.

BOOK IV: SIEGE AND NAVAL WARFARE

125

making arms, iron of both tempers* and coal are kept in magazines.^
Wood, needed for spear-shafts and arrows, is also laid in. Round stones
are very carefully collected from rivers, because they are heavier in
proportion to their density and more suitable for throwing. The walls
and towers are filled with them, the smallest for casting by slings and
“sling-staves”,^ and by hand,"* the larger are shot by mangonels,^ and the
biggest in weight and of a rollable shape are laid out on the battlements,
so that, sent headlong below, they not only overwhelm hostile forces but
smash their machines too. Huge wheels are also made out of green
wood, and cylindrical sections, which they call taleae,^ are cut from
very stout trees and smoothed to make them roll.^ These falling down
with sudden impetus usually deter soldiers and horses. It is also
necessary to have in readiness beams and planks, and iron nails of
different sizes. Besiegers’ machines are usually resisted by means of
other machines, in particular when height needs to be added to walls
and battlements by emergency works, so that the mobile towers of the
enemy may not overtop and take the city.*

9. What to do if the supply of sinews runs out.

It is advisable to collect supplies of sinews^ with the utmost vigour,
because mangonels, catapults and other torsion-engines are of no use
unless strung with ropes made of sinew, although horsehair taken from
the tail and mane is said to be useful for catapults. Indubitably,
however, women’s hair has no less virtue in such kinds of torsion-
engines, as was demonstrated in Rome’s hour of need. For at the siege
of the Capitol, the torsion-engines broke down from continuous and
long fatigue after supplies of sinews ran out. But the matrons cut off

'i.e. iron and steel; it remains unclear whether Roman blacksmiths knowingly added
carbon to harden iron into steel, although they managed to produce it.

^condita, cf. CTh 1A3.

^Justibali, cf III.14.

'‘Cf 1.16. Cf saxorum manualium, “stones for throwing by hand”, Amm. 24.2.14, Livy
38.21.6.

^onagri, “wild-asses”, cf IV.22.

'"“grafts” or “sticks”.

^Described by Athen. Mech. 37.3 as ct({)6v6uXoi or “vertebrae”, perhaps the
cylindh mentioned by Amm. 31.15.13.

“i.e. by means of hoarding, cf IV.17. Also see generally lV.22-23, and D. Baatz, “Town
walls and defensive weapons”, Mavors XI (Stuttgart 1994) 86-90.

'Wrv/.

126

VEGETIUS’ EPITOME OF MILITARY SCIENCE

their hair and presented it to their husbands as they fought, the machines
were repaired and they repelled the hostile attack. For women of the
highest character preferred to disfigure their heads for the moment in
order to live in freedom with their husbands, than become slaves to the
enemy with their beauty intact.' It is also advantageous to collect horns
and raw hides for weaving cataphracts and other machines and armour.^

10. What to do to ensure that the besieged suffer no shortage of
water.

It is a great advantage for a city when its wall includes perennial springs
within it. But if nature does not provide, wells have to be dug to
whatever depth and draughts of water extracted by rope. Sometimes
places are too dry if built upon mountains and rocks. Fortifications
erected on such places may discover sources of water outside the wall,
lower down, and protect them with missiles shot from battlements and
towers so as to allow free access to waterbearers. But if a source is
beyond the range of missiles but on the hill below the city, it is advised
to build a small fortification which they call a burgus^ between the city
and the spring, and station there catapults and archers to defend the
water from the enemy. Also, under all public and many private
buildings cisterns should be constructed very assiduously to provide
reservoirs for rainwater that flows off roofs. It has proved difficult to
overcome with thirst those besieged who used water only for drinking,
however little there was.

'Cf. Serv. ad Verg. Aen. 1.720, as an explanation of the name Venus Calva. The Roman
matron who set the example which V. praises was Domitia. On women’s hair used in
catapults, cf. Flor. 1.31.10 at Carthage c. 149 BC, Caes. Bell Civ. 3.9 at Salonae 49-48
BC. On the siege of the Capitol, cf IV praef note.

^Cf Tac. Hist. 1.79.3: ferrets lamminis aut praeduro corio consertum (of Sarmatian
cataphract contati), “woven with iron scales or very tough leather.”

^According to Kempf, 364, adopted by Roman soldiers from Germans in the first to
second centuries AD. So Emout-Meillet, 78, s.v,, “mot 6videmment germanique”, citing
Orosius 7.32.12: crebraper lirAitem habitaculaconstitutaburgos vocant (sc. Burgundiones
qui inde dicti putantur). “(The Burgundians, thought to be named therefrom), call burgi
die dwelling-places built frequently along the frontier.” According to Waide—Hofmann,
124, s.v., originally borrowed from Greek 7tv3pYO(; = “small fort, watch-tower”, but later
took on the meaning and (e.g. in Sidon. Apoll.) feminine gender of Germanic *burgs =
“fortified settlement”. On boA wells and burgi, cf D. Baatz, Mavors XI (Stuttgart 1994)
79-85.

BOOK IV: SIEGE AND NAVAL WARFARE

127

11. If salt runs out.

If a city is on the coast and salt runs out, water is taken from the sea
and poured into troughs and other flat vessels; in the heat of the sun it
hardens to salt. But if the enemy denies access to the sea, as does
happen, people sometimes collect up sand which the sea stirred up by
the wind has flowed over. They then wash it in fresh water, and this
evaporated in the sun equally well turns to salt.

(Siege-strategies of attack and defence, ch. 12-30)

12. What to do when the first assault is pressed to the walls.

When however a violent assault is prepared against forts and cities,
deadly battles are fought with mutual danger to both sides but greater
bloodshed for the assailants.' For the side wishing to enter the walls
doubles the sense of panic in hopes of forcing a surrender by parading
its forces equipped with terrible apparatus in a confused uproar of
trumpets and men.^ Then because fear is more devastating to the
inexperienced, while the townspeople are stupefied by the first assault
if unfamiliar with the experience of danger, ladders are put up and the
city invaded. But if the first attack is repelled by men of courage or by
soldiers, the boldness of the besieged grows at once and the war is
fought no longer by terror but by energy and skill.^

13. List of machines used in storming walls.

Machines moved up are “tortoises”, rams, siege-hooks, penthouses,
screens, mantelets and mobile towers. For each of them I shall describe
how they are constructed, and how they are used in attack or repelled.

14. The ram, siege-hook and “tortoise”.

The “tortoise”"^ is made from timbers and planks with a covering of
hides, goafs hair mats^ and fire-blankets^ to save it from destruction by

’Cf. Amm. 19.9.9 on Persian losses at Amida, Nisibis and Singara, ad 359. Cf. Onas.
42.7-13, Livy 37.5.5, Tac. Hist. 5.13.4. This is the second kind of siege warfare
foreshadowed at IV.7 init

^e.g. by king Sapor (Shapur) 11 at Amida, Amm. 19.2.1 ff., and by Constantins 11 at
Bezabde, Amm. 20.11.8.

'Cf. Livy 38.5.3-4.

^testudo, siege-shed.

Cilicia., cf. IV.6.

128

VEGETIUS’ EPITOME OF MILITARY SCIENCE

fire.' It holds within it a beam which is either tipped with an iron hook
and called a falx^ because it is curved for tearing stones out of the wall,
or else the head of the beam itself is covered in iron and called a
“ram”.^ This is either because it has an extremely hard brow for
undermining walls, or because it backs off like a ram in order to strike
harder at speed. The “tortoise” takes its name from its resemblance to
the real tortoise. Just as it now withdraws, now thrusts out its head, so
the machine at one moment withdraws its beam, at another sends it out
to strike more strongly."

15. “Vines'’, screens and the siege-mound.

The ancients called “vines’’^ what are now called in military and
barbarian parlance cau<s>iae.^ The machine is made of light wood,
8 ft. wide, 7 ft. tall and 16 ft. long. The roof is constructed with a
double protective covering of boards and hurdles. The sides also are
fenced with wicker against penetration by impact of stones and missiles.
To avoid combustion from fire-darts the outside is covered with raw and
freshly flayed hides^ and fire-blankets.* When a number have been
made, they are joined together in a line,^ and under their shelter

^'centones, cf. centonarius, a “fireman who used mats for extinguishing fires”, OLD s.v.
‘Cf. Caes. Bell. Civ. 2.9.4, 2.10.6.

^literally “sickle”, i.e. siege-hook.

^aries. Cf. the Roman ram with its shed-timbers covered against fire-attack with wet
hides, fire-blankets and alum at Bezabde, ad 360, Amm. 20.11.13, and the Ostrogothic
rams used in the siege of Rome ad 536-537, Procop. Bell. Goth. 1.21.5-12.

'*Cf. Varro de ling. lat. 5.161: testudo dicebatur ab testudinis similitudine, ut est in
praetorio et castris, “‘tortoise’ was said after its likeness to the tortoise, as it is in the
military headquarters and in the camp”; cf Jos. Bell. Jud. 3.213-218.

\ineae, penthouses.

^^So Lang, meaning literally a kind of Macedonian hat, but the word is garbled in the
MSS; cautiae, cauciae and cautib- are the best evidence we have. Kempf, 365 n.3,
suggested cautia may be Germanic, related to Anglo-Saxon cyte or cote, cottage. See O.
Lendle, Texts und Untersuchungen zum technischen Bereich der Antiken Poliorketik,
Palingenesia XIX (Wiesbaden 1983) 139-141.

’Cf Jos. Bell. lud 3.173.

“Cf IV. 14.

'M.e. in a line roughly perpendicular to the wall; cf Caes. Bell. Civ. 2.10.1, Livy 21.7.5.
They were also used with rams (ibid. 21.7.5), mobile towers (8.16.8) and siege-mounds
(37.26.8).

BOOK IV: SIEGE AND NAVAL WARFARE

129

besiegers make openings to undermine the foundations of walls in
safety.*

“Screens”^ refers to apse-shaped structures made from wicker, covered
with goat’s hair mats and hides, and fitted with three wheels, one in the
middle and two at the ends, so that they can be moved up in whatever
direction you wish, like a wagon. Besiegers bring them up to walls and
sheltering under their cover dislodge all defenders from fortifications by
means of arrows, slings and missiles, so as to provide easier opportunity
to mount by ladders.^

The siege-mound"* is built from earth and timbers against the wall,
and from it missiles are shot.^

16. Mantelets.^

They call musculf the smaller machines which shelter soldiers who
remove a city’s stockade;* they also fill the fossework^ with stones,
timber and earth that they bring up, even making it solid so that mobile
towers can be joined to the walls without obstruction. Musculi are
named after the sea-creatures; for as they, though quite small, provide
continuous support and assistance to whales,'** so these diminutive

'Cf. Livy 23.18.8, 38.7.6; and cf. Amm. 24.4.13, 21-22, Zos. 3.22.2 for the same at
Maozamalcha ad 363.

^plutei^ cf. Amm. 21.12.6 at Aquileia AD 361, where the pluteus is indistinguishable from
the vinea and is used to cover sappers. The triangular chassis with three wheels which are
illustrated on Trajan’s column, Cichorius PI, LXXXV scene cxiv, may be intended to
represent plutei.

^Also described by Athen. Mech. 38.9-13. The tichodifrus of the Anon, de Rebus Bellicis
8, 19.5-6, resembles it in size and purpose. See Lendle, 144-146.

^agger.

^Cf Amm. 19.8.1-2 at Amida ad 359, Amm. 20.11.12 at Bezabde AD 360; Procop. Bell.
2.26.29, Maur. Strat. 10.1.14 know it as an dyeaxa, derived from the Latin aggestus^
a term also used by Amm.

^’Cf. Caes. Bell. Civ. 2.10-11, a detailed description of a musculus indistinguishable from
a vinea, and used to cover sappers. Cf Athen. Mech. 15.12 ff See Lendle, 141-144.

^“mice, mussels, muscles”, i.e. mantelets.

^sudatum, cf Maur. Sirat. 12.8.22. This is the earliest occurrence of sudaturn in Latin
literature. Cf F. Lammert, “Suda, die Kriegsschriftsteller und Suidas”, Byz.Z. xxxviii
(1938) 27.

yossatum, cf SHA Gord. 28.3, CTh 7.15.1. Cf F. Lammert, loc. cit. For the process of
levelling ditches by filling with faggots, cf Procop. Bell. Goth. 1.21.13.

'"Cf Pliny Hist. Nat. 9.186, 11.165, Claud. In Eutrop. 2.425-431, apparently describing
pilot-fish, and certainly not what we call “mussels”.

130

VEGETIUS’ EPITOME OF MILITARY SCIENCE

machines are assigned to the big towers and prepare the way for their
advance, building roads ahead.*

17. Mobile towers.^

“Towers”^ refers to machines constructed from beams and planks
looking like buildings. They are very thoroughly armoured with raw
hides and fire-blankets, lest all this work be burnt by enemy fire. Their
width increases in proportion to the height, for sometimes their
dimensions are 30 ft. square, sometimes 40 ft. or 50 ft." Their height is
sufficient to overtop not only walls but even towers built of stone. Many
wheels are placed under them by mechanical skill, so that by their
rolling motion such a great bulk may be moved. The danger to a city is
immediate once a mobile tower is moved up to the wall. For it holds
within it many ladders, and attempts to make a breach in different ways.
In the lower storey it contains a ram, whose impact destroys walls.
About the middle storey it holds a bridge, made from two beams and
fenced with wicker.^ This is suddenly thrust out between the tower and
wall and made secure, and soldiers emerging from the machine cross
over by it into the city and occupy the walls. In the upper storey of the
tower are stationed pikemen and archers, who cut down the defenders
of the city from above with pikes, javelins and stones.^ When this
happens the city is captured without delay. What help is there left when
those who were counting on the height of their walls suddenly spy a
higher wall of the enemy above their heads?^

'The colourful etymology is not attested elsewhere, and may well be V.’s. More likely
the machine is named after the diminutive of mus or mouse, whose shape it vaguely
resembled.

^turres ambulatoriae, see Lendle, 101-105.

^turres.

Witr. 10.16.4 says that the breadth of the famous helepolis of Demetrius was 60 ft., and
the height 125 ft.

^Cf. IV.21, where it is identified as the exostra, “thrust-bridge”.

^’At Julian’s siege of Aquileia AD 361, three mobile towers were constructed with a
drawbridge in the lower storey and soldiers shooting from the upper, and were conveyed
to the city-walls on rafts, cf Amm. 21.12.9-10. The Ostrogothic king Vittigis built mobile
towers drawn on wheels by oxen for the siege of Rome ad 536-537, cf Procop. Bell.
Goth. 1.21.3-4.

^At the siege of Pirisabora AD 363, the sight of Julian’s helepolis under construction was
enough to induce the besieged to offer acceptable terms of surrender, cf Amm. 24.2.18 ff.

BOOK IV: SIEGE AND NAVAL WARFARE

131

18. How a mobile tower may be fired.

This very obvious peril is averted by a number of methods. First, if
there are brave men or a force of soldiers on hand, a sortie is made in
which a group' of armed men goes out and, violently repelling the
enemy, pulls the hides off the woodwork and burns the huge machine.
But if the townspeople dare not go out, they shoot from larger catapults^
lighted fire-darts and fire-spears, so that piercing through the hides and
fire-blankets the flame may be planted inside it. Fire-darts^ are like
arrows, and wherever they lodge they bum everything, because they
come blazing."* The flaming spear^ is like a spear in that it is fitted with
a strong iron head; between the tube and the spearshaft is a wrapping
of sulphur, resin, bitumen and tow, steeped in the oil which they call
“buming-oil”.^ It is sent by the force of the catapult, pierces through the
armour, is fixed blazing in the wood, and often burns up the tower¬
shaped machine.’ Men are also let down on ropes while the enemy are
asleep; carrying lights in lanterns they set fire to the machines and are
hoisted up onto the wall again.^

19. How height is added to the walls.

Besides this the part of the wall which the machine attempts to reach is
made higher by building it up with cement and stones, or mud or bricks,
and finally with hoarding,^ so that the machine cannot destroy the
defenders of the walls by attacking the city from above.'® Naturally the

^globus^ platoon.

^ad maiores ballistas: V. occasionally lapses into late>Latiii instrumental ad, cf. 11.15 n.
^malleoli.

'Cf. Amm. 23.4.14-15, 20.11.13, 21.12.10, 24.4.16; Eusebius 2.2-4, F. Jacoby

(ed.), FGH IIA p.480 (Gothic siege of Thessalonica, probably ad 253 or 269); Paul. Fest.
135 M.; Sisenna Hist. IV fr. 83; see S. James, "‘Archaeological evidence for Roman
Incendiary Projectiles”, 51/xxxix (1983) 142-143 with further refs.

^falarica.

oleum incendiarium, cf IV.8. Amm. 23.631 describes “Medic oil", a Persian product
and perhaps a kind of petroleum, as used in fire-darts.

^Cf. Livy 21.8.10-12.

’*Cf. Dexippus Scyth. fr. 27.9, F. Jacoby (ed), FGH IIA p.471 (Gothic siege of
Philippopolis, probably ad 250-251), and the Persian sorties at Bezabde ad 360 described
by Amm. 20.11.18, 22-23.

‘^i.e. wooden boarding erected on stone fortifications to screen the defenders. Cf. Caes.
Bell. Gall. 7.22.3.

'‘‘An allusion to Verg. Aen. 2.46-47; aut haec in nostros fabricata est machina muros \
inspectura domos venturaque desuper urbi, “or else this is a machine constructed to harm
our walls, to spy out our homes and to attack the city from above.”

132

VEGETIUS’ EPITOME OF MILITARY SCIENCE

machine is rendered ineffective if it is found to be lower. But besiegers
are in the habit of using the following kind of stratagem. First they
build a tower in such a way that it looks smaller than the battlements of
the city. Then they secretly make another turret inside out of planks and,
when the machine is joined to the walls, suddenly the turret is pulled up
from the middle on ropes and pulleys. From it emerge soldiers who,
because it is found to be higher, at once capture the city.'

20. How the ground may be undermined so that the machine can be
rendered harmless.

Sometimes very long iron-clad beams are opposed to an approaching
machine and push it away from the vicinity of the wall. But when the
city of Rhodes was under attack by enemies and a mobile tower higher
than all the walls and towers was in preparation, the following remedy
was invented by the genius of an engineer. During the night he dug a
sap under the foundations of the wall and, removing the earth without
any of the enemy realizing, hollowed out inside the place to which the
tower was to be advanced on the following day. When the mass was
moved on its wheels and reached the place that had been undermined,
the soil gave way under such great weight and it subsided and could not
be joined to the walls or moved farther. So the city was liberated and
the machine abeindoned.^

21. Ladders, drawbridge, ‘‘thrust-bridge” and swing-beam.

When mobile towers have been moved up, men are cleared from the
walls by the action of slingers with stones, archers with darts, hand-
catapultiers^ and crossbowmen"^ with arrows, and darters^ with lead-
weighted darts^ and javelins. This done, they put up ladders’ and occupy

'Cf. Lendle, 105-106.

^The siege of Rhodes was conducted by Demetrius I Poliorcetes, in 305-304 BC. The
tower was the famous helepolis, cf Diod. Sic. 20.91.2 ff.. Pint. Dem. 21, Athen. Mech.
27.2 ff., Vitruv. 10.16.4-8, Amm. 23.4.10. V. diverges from the version of the story in
Vitruvius in which the Rhodians drenched the ground in front of the helepolis with
sewage, so that it sank into the mire when moved forwards. V.’s tactic is that
recommended by Aen. Tact. 32.8, writing earlier in the fourth century BC (cf D.
Whitehead, 8-9).

^manuballistaniy cf IV.22.

*arcubailistarii, cf IV.22.

^iaculatores.

^'plumbatae, cf 1.17,

BOOK IV: SIEGE AND NAVAL WARFARE

133

the city. But those climbing up ladders are often put in peril. For
example, Capaneus—reputedly the first to discover storming by
escalade—^was slain with such violence by the Thebans that it was said
he had been blasted by a thunderbolt.'

For this reason besiegers get across onto the wall of the enemy by
means of the drawbridge, “thrust-bridge” and swing-beam. The
sambucd^ is named after its likeness to the zither. For corresponding to
the strings on the zither, there are ropes on the beam which is placed on
the side of the mobile tower, and these let the drawbridge down from
the upper storey on pulleys, to descend onto the wall.^ Warriors
immediately exit from the tower and cross over by it to invade the walls
of the city. The exostra^ denotes the bridge we described earlier,^
because it is suddenly thrust out onto the wall from the tower.^ The
tolleno' is the term for a very tall pole planted in the ground, which has
attached to its top end a cross-beam of longer dimensions, balanced at
the middle so that if you depress one end, the other is raised. On one
end is constructed a machine made from hurdles and boards, and in it
a few soldiers are placed. Then as the other end is pulled and lowered
by ropes, they are lifted up and deposited on the wall,®

22. Catapults, mangonels, ^^scorpions’% crossbows, “sling-staves’* and
slings, torsion-engines by which the wall is defended.

The besieged normally defend themselves against these contrivances
using catapults, mangonels, “scorpions”, crossbows, “sling-staves” and
slings. The catapult^ is strung with ropes of sinew, and the longer arms

^scalae.

'V. offers a Christian(?) rationalizing account of the story from the mythical war of the
Seven against Thebes which is the central episode of Statius’ Thebaid at 10.837-939, in
which Jupiter kills Capaneus with a thunderbolt

^literally “lute”, i.e. drawbridge.

^See Lendle, 104. Cf. Polyb. 8.4.11 and Lendle, 176 on the etymology. There is only a
vague resemblance to a musical instrument

“‘thruster-out” in Greek, i.e. “thrust-bridge”.

’Cf IV.17.

‘^See Lendle, 101-104.

’swing-beam.

*See Lendle, 117-127, cf Polyb. 8.5.8-11, 8.6.1-4, Livy 24.34.10-11 (used to lift ships
up on end), 38.5.4 (used to drop weights onto rams).

^hallista.

134

VEGETIUS’ EPITOME OF MILITARY SCIENCE

it has, that is, the bigger the machine is, the farther it shoots darts.* If
it is tuned in accordance with mechanical art and aimed by trained men
who have worked out its range in advance, it penetrates whatever it hits.
But the mangoneP shoots stones, and throws various weights in
proportion to the thickness of the sinews and size of stones. The larger
the machine, the bigger the stones it hurls like a thunderbolt.^ No type
of torsion engine more powerful than these two types is found.^

TTiey used to call “scorpions” what are now called manuballistae;^
they were so named because they inflict death with tiny, thin darts.® I
think it superfluous to describe “sling-staves”,^ crossbows® and slings,^
which are familiar from present-day use.*° Heavy stones thrown by the
mangonel destroy not only horses and men, but also the machines of the
enemy.

^spicula.

^onager, literally “wild ass”.

’Cf. IV.29.

'Cf. Amm. 23.4.1-7, Procop. Bell Goth. 1.21.14-19. Note that V., like all authors of the
late Empire, uses ballista for an arrowTirer, which would have been called a catapulta
in the early Empire, and onager for a stone-thrower, which would have been called a
ballista, no doubt because of technological changes—on which see E.W. Marsden, Greek
and Roman Artillery: Historical Development, 188 ff., id. Greek and Roman Artillery:
Technical Treatises, ch. VI I-VIII.

^hand-catapults.

^'spicula. Amm. 23.4.7 says scorpio was the old name for the onager. D.B. Campbell,
“Auxiliary Artillery Revisited”, A/clxxxvi (1986) 128 n.82, listing earlier ancient sources
which apply the term exclusively to static arrow-firing field-artillery, says that Ammianus
and V. are both wrong. The manuballista is probably (Ps.-)Heron’s Cheiroballistra, on
which see E.W. Marsden, Tech. Tr. 236,209, D. Baatz, “Recent finds of ancient artillery”,
Britannia ix (1978), 1-17, = Mavors XI224—245, Campbell, 130. Torsion-engines of diis
type, though much larger, first appear on Trajan’s column, Cichorius PI. XXXI scene xl,
PI. XLVII scene Ixvi.

^Justibali.

*arcuballistae.

^jitndae.

'®V. forgets that he earlier (at 111,14 fin.) described the fustibalus and funda. The
arcuballista was invented by the early Empire, at least, cf. Arr. Tact. 43.1, D.B.
Campbell, art. cit., 131-132, and the manuballista may have differed from it principally
in the means of propulsion, using torsion not tension. Such at any rate seems to have
determined the friirteenth-century illuminator of MS Marley Add.l Vegetius De Re
Militari, f,86 (Fitzwilliam Mus., Cambridge), to draw one “crossbow” with a thick, twisted
sinew, and the other with a fine bowstring. See also D. Baatz, art. cit., 14-15.

BOOK IV: SIEGE AND NAVAL WARFARE

135

23. Mattresses, nooses, grapnels and heavy columns are useful
against rams.

There are also a number of remedies against rams and siege-hooks.
Some people let down' on ropes quilted blankets^ and mattresses,^
putting them in front of the places where the ram strikes, so that the
impact of the machine, weakened by the soft material, may not destroy
the walL* Others catch the rams in nooses^ and, using gangs of men,
drag them from the wall [up]^ at an angle, overturning them sheds and
all.’ Many attach ropes to a toothed iron instrument like a pair of
pincers, which they call a “wolf’,* and catching the ram they either
overturn it or they hang it up so it loses the impetus to strike.^
Sometimes marble bases and columns launched with great impetus from
walls smash rams.'® But if such force is used that the wall is pierced by
rams and, as often happens, it falls down, one hope of safety remains,
and that is to demolish houses and build another wall inside." The
enemy may then be wiped out between the two walls if they attempt to
enter.

24. On saps, whereby the wall is undermined or the city
penetrated.*^

Another method of assault is subterranean and secret, which they call a
“burrow”,'^ after the hares which dig tunnels underground and hide in
them. A gang is set to work digging the earth with great labour as in the
mines where the industrious Bessi explore veins of gold and silver,"* and
by means of the excavated cave an underground route to the city’s

^Chalo, -are, a graecism, already in Vitr. 10.8.1; cf. chalatorios (1V.46).

^centones.

^culcitae.

'*Cf. Aen. Tact. 32.3, bales of chaff, woolsacks, freshly-flayed oxhides inflated or stuffed,
Jos. Bell. lud. 3.223, bales of chaff.

^laquei, cf III.23 where it means “lassos”.

insert the reading sursum, “up”, from MS n, which seems to explain in obliquum.
"Cf Aen. Tact. 32.4, Thuc. 2.76.4, Caes. Bell. Gall 7.22.2, Amm. 20.11.15.

*lupus, i.e. grapnel.

"Cf Lendle, 194-196; cf Livy 28.3.7.

"'Cf Amm. 20.11.10, 31.15.14, Tac. Hist. 2.22.2.

’‘e.g. Thuc. 2.76.3, at Plataea, 429 BC., and Diod. Sic. 20.93.1, 97.4, at Rhodes, 304 BC.
‘^Cf Aen. Tact. 37.

^^cuniculus, i.e. sap.

''Cf 11.11.

136

VEGETIUS’ EPITOME OF MILITARY SCIENCE

destruction is sought.' This stratagem is effected by two methods of
attack. Either they penetrate beneath the city and, emerging by the
burrow at night unbeknown to the townspeople, open the gates to admit
a column of their own side who kill the enemy in their own homes
taking them unawares;^ or else when they reach the foundations of the
walls, they excavate the largest possible part of them, placing dry timber
there and holding up the collapse of the wall by temporary works. They
also add brushwood and other inflammable tinder. Then, when the
soldiers are ready, fire is introduced to the work and, all the wooden
props and boards having burned, the wall suddenly collapses, opening
a way for invasive action.

25. What the besieged should do if the enemy break into the city.^
Countless examples demonstrate that enemies have often been slain to
a man after they had invaded a city. This is the certain result, if the
citizens hold on to the walls and towers and occupy the higher ground.
For then from windows and rooftops people of all ages and both sexes
overwhelm the invaders with stones and other kinds of missiles. To
avoid this fate, the besiegers frequently open the city gates in order to
induce resistance to stop by conceding the chance to escape. For
necessity is a desperate kind of courage."^ In this case there is only one
help for the besieged, whether the enemy enters by day or night, and
that is to hold the walls and towers, climb the higher ground, and
overwhelm the enemy from all sides by fighting it out in the streets and
squares.^

26. What precautions should be taken to prevent the enemy
occupying the wall by stratagem.

Besiegers often think up a stratagem in which they take themselves far
off in feigned despair. But when all fear has gone and the town relaxes

'Cf. Caes. Bell. Gall. 7.22.2, linking Gallic skill at saps with their iron-mining.

^Cf Amm. 24.4.21-23, Zos. 3.22, siege of Maozamalcha, ad 363.

'Cf. Onas. 42.18-22.

"Cf ni.21.

^Cf Front. Strat. 2.6.9: Pyrrhus Epirotarum rex, cum quandam civitatem cepisset
clausisque portis ex ultima necessitate fortiter dim icontes eos, qui inclusi erant,
animadvertisset, locum illis ad fugam dedit. “When Pyrrhus, king of the Epirotes, had
taken some city and, having closed the gates, noticed that those shut inside were fighting
strongly on account of their extreme peril, he gave them the chance to escape.” Cf
Polyaen. 3.9.3, Leo Tact. 15.21.

BOOK IV: SIEGE AND NAVAL WARFARE

137

into careless complacency, abandoning the watches on the walls, they
seize their chance in the darkness of night to come stealthily with
ladders and climb the walls. Against this increased security should be
used when the enemy withdraws, and small huts’ should be placed on
the walls and towers themselves, in which sentries may be protected in
winter months from rain and cold, in summer from sun,^ Experience
also discovered the trick of keeping very keen-scented and alert dogs in
towers, which can anticipate the arrival of the enemy by their scent and
reveal it by barking.^ Geese also by their clamour indicate night-attacks
with equal skill. For having attacked the Capitoline citadel the Gauls
would have destroyed the the very name of Rome, had not Mallius been
roused by the clamour of the geese to stop them. Marvellous was the
watchfulness or good fortune, whereby one bird saved the men destined
to send the whole world under the yoke."*

27. When surprise attacks are made on the besieged.

Not just in sieges but in every kind of warfare it is deemed of the
highest importance to spy out and get to know thoroughly the habits of
the enemy.^ For an opportunity for a surprise attack cannot otherwise be
found unless you know the times when the adversary leaves off
attending to his work, when he is rendered less careful, sometimes at
mid-day, sometimes in the evening, frequently at night, otherwise at the
meal-times, when soldiers on both sides disperse to rest or look after the
body. When this starts to happen in a city, the besiegers cunningly
remove themselves from the action to give free rein to their adversaries’
negligence. When this has increased as it goes unpunished, suddenly
they move up machines or put up ladders and capture the city.^ For that
reason stones and the other torsion artillery are placed in readiness on
the walls,^ so that when a surprise attack is detected those running to

^teguriola, also spelled tuguriola, n class MSS, and Apiil. Met. 4.12.

'Cf 11.23.

^e.g. at the siege of the Capitol, 390 BC, cf Livy 5.47.3.

'’M. Mallius (Capitolinas), hero of the siege of the Capitol. For the story, cf Livy 5.47,
Florus 1.7.15, Ambrose Ep. 18.4. V. aligns himself with the Christian rationalizing
tradition. Cf also Ovid Met 11.599: sollicitive canes canibusve sagacior anser, “nervous
dogs or, more alert than dogs, a goose.”

'Cf. 111.6.

^‘Repeats IV.26, a sign that V. is elaborating on jejune source-material.

'Cf. IV.8.

138

VEGETIUS’ EPITOME OF MILITARY SCIENCE

resist may have to hand the material to roll and shoot down onto the
heads of the enemy.

28. What the besiegers do to avoid suffering surprise attacks from
the besieged.'

When negligence intervenes, besiegers are equally subject to surprise
attack. For whether they are preoccupied with food or sleep, or
dispersed to rest or on some other necessity, then is the time when the
citizens suddenly break out and kill them unawares, setting fire to the
rams, machines and even the siege-mounds, and overthrowing all the
works constructed for their own ruin.^ Against this the besiegers make
a fosse beyond the range of missiles, and equip it with a rampart,
stockade and turrets, so that they can resist sorties from the city; they
call this work a “breastwork”.^ Often in descriptions of sieges in
historical works one finds that a city has been surrounded with a
breastwork."*

29. The type of torsion-engines with which a city is defended.
Spears, lead-weighted darts, lances, longer Javelins,^ and shorter
Javelins^ fall more violently on those below when aimed from a height.
Also arrows shot from bows and stones thrown by hand, slings or
“sling-staves” penetrate farther the higher the point of origin.’ But
catapults and mangonels, provided they are tuned very carefully by
experts, surpass everything else. No amount of courage or armour can
defend soldiers from them. For like a thunderbolt, they generally either
smash or pierce whatever they hit.*

‘Cf Onas. 40, which may well have influenced Frontinus’ De Re Militari, and hence
indirectly V.

^e.g. Jos. Bell. lud. 3.227-228, cf. also Livy 5.7.2.

^loricula.

^Cf. Polyb. 9.41.1-3. The term is usually lorica in Latin, cf. Caes. Bell. Gall. 5.40.6,
Florus 1.34.13, Amm. 24.5.2, 31.3.7, 31.15.4; but the TLL cites for the diminutive
loricula^ as well as V., A. Hirtius Bell. Gall. 8.9.3, Itala Biblia Dew/. 22.8, and Hieron.
in Is. 26.1.

\eruta.

^'spicula.

’Cf. 111.13.

“Repeats IV.22. Cf the ballista fulminalis —“thunderbolt” catapult—described by the
Anon, de Rebus Bellicis 18 as able to shoot over the Danube. Jos. Bell. lud. 3.243 ff.
gives a vivid account of the power of torsion-engines in attacking cities, as does Amm.
19.5.6, 19.7.4-7, 31.15.12, in defending them. Cf Amm. 24.4.28 for the gruesome death

BOOK IV: SIEGE AND NAVAL WARFARE

139

30. How the measurement is obtained for making ladders and
machines.

Ladders and machines are most useful for capturing walls if they are
made to such a size that they surpass the height of the city.’ The
measurement is worked out by two methods. Either a thin, light thread
is tied at one end to an arrow, and when it reaches its mark having been
aimed at the top of the wall, the height of the walls is found from the
length of thread. Or else when the slanting sun casts a shadow of the
towers and walls on the ground, the length of the shadow is measured
without the knowledge of the enemy. At the same time a ten-foot rod
is fixed upright in the ground and its shadow measured in the same way.
With this information, the height of the city is undoubtedly revealed by
the shadow of the ten-foot rod, as it is known what height casts what
length of shadow.^

Both the recommendations of the authors of arts of war for attacking
and defending cities and what the experience of recent emergencies has
discovered, I have summarized for the public benefit, as I believe,
making the point again and again that the most thorough precautions
should be taken against shortage of drinking-water or food arising at any
time, since such troubles cannot be ameliorated by any stratagem.
Therefore, so much the more should be stockpiled within walls, in
proportion to the length of blockade known to be within the capability
of the besiegers.^

of an engineer— architectus —caused by the misfiring of one of his own mangonels.

'Cf Amm. 20.11.20 (siege-mounds), 21.12.6 (ladders). Procop. Bell. Goth. 1.21.5
(ladders).

^Cf. Polyb. 9.19.5-20.4; Jul. Afr. Cest. 1.15 recommends using a dioptra or alidade on
the end of the measuring-rod. For counting courses of bricks or stones to calculate the
height of walls, cf. Thuc. 3.20.3-4, Procop. Bell. Goth 1.21.3-4.

W. singles out IV.7 for special emphasis, as it was particularly relevant to contemporary
conditions of barbarian invasion and attacks on civilian communities. It is assumed that
the Roman army was not going to be able to lift every blockade.

140 VEGETIUS’ EPITOME OF MILITARY SCIENCE

(The Navy, ch. 31-33)

31. Precepts of naval warfare.*

Now that by Your Majesty’s command, Invincible Emperor, the
accounts of land warfare have been cleared, the balance outstanding
belongs in my opinion to naval warfare. Its arts require less to be said
for the reason that the sea has long been pacified, and our struggle with
barbarian races is played out on land.^

The Roman People for the pomp and advantage of their Empire used
not to fit out the fleet on the spur of the moment in response to the
needs of some crisis, but always kept it in readiness lest they should
ever be in danger.^ For no one dares to challenge to war or inflict injury
upon a kingdom or people he knows is armed and ready to resist and
avenge any attack."* So one legion was stationed with each fleet at
Misenum and Ravenna, not too far away to protect the City and yet,
when policy required, able to reach by navigation without delay or
detour all parts of the world.^ For the fleet at Misenum had nearest to
it Gaul, the Spains, Mauretania, Africa, Egypt, Sardinia and Sicily. The
fleet at Ravenna was used to reach by direct sailing Epiros, Macedonia,
Achaia, the Propontis, Pontus, Oriens, Crete and Cyprus.^ For in matters
of war speed is often more useful than courage.

* Aen. Tact. 40.8 appended a treatment of naval warfare, lost; so too his successors in the
Greek tactical tradition, Philo Byz. Potior. 104, cf. Ael. Tact. 2.1, Asclep. 1.1. See F.
Lammert, "‘Die Mteste erhaltene Schrift Uber Seetaktik und ihre Beziehung zum Anonymus
Byzantinus des 6. Jahrhunderts, zu Vegetius und zu Aineias’ Strategika'\ Klio xxxiii
(1940) 280-282. It is included here not because V. wanted to write a complete art of war
(he did not), but because he wanted to correct neglect of the navy by the Government.

^Points to a date at least before the Vandal naval capability which was already
considerable by ad 419; cf J.R. Moss, “The effects of the policies of Aetius on the
history of western Europe”, Historia xxii (1973) 723-728, CTh 9.40.24 (419): His, qui
conficiendi naves incognitam ante peritiam bar bar is tradiderunt... capitate... suppticium
proponi decernimus. “For those who have handed over to the barbarians previously
unknown knowledge of how to build ships... we decree the imposition of capital
punishment.”

^Cf the principle of Roman military readiness enunciated also by Jos. Bett. lud. 'i.ll.
*A restatement of the famous dictum si vis pacem para bettum, cf III praef

^Suet. Aug. 49.1: classem Miseni et atteram Ravennae ad tutetam Superi et Inferi marts
conlocavit... “(Augustus) stationed one fleet at Misenum and the other at Ravenna, for the
protection of the Upper (Adriatic) and Lower (Tyrrhenian) seas.”

*The list is V.’s own composition; Egypt was part of Oriens until c. ad 367, cf Jones,
141.

BOOK IV: SIEGE AND NAVAL WARFARE

141

32. The titles of the officers commanding the fleet.

The prefect of the fleet at Misenum was in command of the warships’
stationed in Campania, whilst those located on the Ionian Sea were
controlled by the prefect of the fleet at Ravenna. Under each of them
were ten tribunes appointed one for each cohort. Each warship had a
single navarch,^ that is, a kind of skipper,^ who was exempted"* from the
other duties of sailors and put in a daily responsibility and unfailing
efforts to training pilots,^ oarsmen^ and marines.^

33. How warships got their name.”

Different provinces at various times held considerable naval power and
therefore the types of ships were diverse. But when Augustus was
fighting at the battle of Actium^ and Antony was beaten mainly by the
auxiliaries provided by the Libumi, it became clear from the experience
of that great encounter that the ships of the Libumi were better-designed
than the rest. Therefore usurping the likeness and the name, the
Emperors built the fleet according to their pattern. Liburnia is a part of
Dalmatia lying next to the city of Iadera;‘° ships of war are built today

Uiburnae.

^navarchus. The captain of the ship, he is probably seen by V. as ranking with the
centurion in the land legion, cf. Grosse, 116.

^naviculahus, technically a merchant shipowner.

V.l. excepti (pro exceptis), cf. A. Eussner, “Zu Vegetius”, Philologus xliv (1885) 87.
^gubernatores.

^'remiges.

^milites. Note that there is no evidence that the navarchus was the commander of a
Roman naval vessel; this was in fact the therarchus. But the navarchus was at any rate
senior to the trierarchus on the same ladder of promotion. V.’s equation with a
navicularius is misleading since the latter did not necessarily go to sea. Cf. M. Redd6,
Mare Nostrum (Rome 1986) 542. See generally G. Forni, Esercito e Marina de Roma
Antica, Mavors V (Stuttgart 1992) 298-323, 419-450.

**M. Redd6, 105. “Liburnian” was used for all warships of whatever size (cf IV.37) by
the late Empire, not only the light biremes of the first century BC which gave the name.

'^2nd. Sept., 31 BC, off the west coast of Greece; Augustus as C. Julius Caesar
(Octavianus) divi filius with his admiral M. Vipsanius Agrippa decisively defeated M.
Antonius and Cleopatra and took sole control of the Roman world.

"’Zadar.

142

VEGETIUS’ EPITOME OF MILITARY SCIENCE

on their model and are called liburnae.^

(Shipbuilding, ch, 34-37)

34. The care with which warships are built.

As when building houses the quality of the sand or stone of the
foundations is important, so the more carefully should all materials be
obtained when building ships, because it is more dangerous for a ship
to be faulty than a house.^ So the warship is constructed principally
from cypress, domestic or wild pine, larch,^ and fir/ It is better to fasten
it with bronze nails than iron; for although the cost seems somewhat
heavier, it is proved to be worthwhile because it lasts longer, since iron
nails are quickly corroded by rust in warm, moist conditions, whereas
bronze preserve their own substance even below the water-line/

35. The astronomical observations according to which timber should
be cut.

Take particular observations so that trees from which warships are to be
built are felled between the 15th moon and the 22nd. On these eight
days alone is cut timber which is preserved immune from decay; that
cut on the other days turns to dust even within the same year, eaten

‘Appian Illyr. 3.7: Kai vauxiKoi' pev xoiq ’Ap6iaioiq ^yevovto Aipepvoi', y^voq
Sxepov M^iupico V, oCxov ’I6viov Kai xdq vi^orouq ^X^ctxeuov vauCTivwKEiaiqxEKai
Koi5(|)aic;, 60£v tn vuv ‘PcDpaioi xd Ko\j(t)a Kai d^ta 5iKpoxa Aipupv^Saq
TrpoaayopEU ouaiv. “Also sailors in addition to the Ardiaeans were the Libumians, another
Illyrian tribe, who used to raid the Ionian sea and islands using swift, light ships; hence
even today the Romans call light, quick biremes ‘libumians’.” Cf. Lucian Am. 6. Only V.
gives the detail about the high performance of Octavian’s Uburnae at the battle of Actium,
31 BC, but the testimony may to some extent be corroborated by that of Cass. Dio
50.18.4 ff. Cf also Hor. C 1.37.30, Epod. 1.1, Prop. 3.11.44, Plut. Ant. 67.2, Prud. c.
Symm. 2.530-531. See W.M. Murray, P.M. Petsas, Octavian's campsite memorialfor the
Action War, Trans, of the American Philosophical Society lxxix.4 (1989) 147 ff., who
argue that the “heavy fleet vs. light fleet” tradition of the battle derives from the
appearance in the 20s BC of Augustus’ own Memoirs.

^Perhaps an inept allusion to Biblia Vulgata Matt. 7.24-27, from the Sermon on the
Mount. For the parallel should look to the quality of what the houses are built of, not
what they are built on.

larice, transmitted in all MSS of e group, omitted by tc.

^See R. Meiggs, Trees and Timber in the Ancient Mediterranean World (Oxford 1982)
86, 118-120.

^Cf Caes. Bell. Gall. 4.31.2: quae gravissime adjlictae erant naves, earum materia atque
aere ad reliquas reftciendas utebantur, “as for the ships that were most seriously
damaged, they used their timber and bronze to repair the rest.”

BOOK IV: SIEGE AND NAVAL WARFARE

143

away by destruction of worms.* This has been the lesson of science
herself and the everyday experience of all shipbuilders, and we
recognize it too when we contemplate the very religious festival which
it has been decided to celebrate for ever more on these days alone.^

36. The months in which timber should be cut.

Timbers are best cut after the summer solstice, that is, through the
months of July and August,^ and through the autumnal equinox until the
1st January. For in these months the moisture evaporates and the wood
is drier and therefore stronger. Avoid sawing timbers immediately after
felling, or putting them into the ship as soon as they have been sawn,
because both trees that are still whole and those divided into “double”
planks'^ deserve a truce for further drying. Those fitted when still green
exude their natural moisture and contract, forming wide cracks; nothing
is more dangerous for sailors than for the planking to split.^

37, The size of warships.

So far as size is concerned, the smallest warships have one rank of oars
a side, those slightly bigger two ranks, those of appropriate dimensions
three, four, sometimes five ranks for their oarage. This should not seem
enormous to anyone, because at the battle of Actium far larger vessels

'So Servius ad Verg. Georg. 1.256: Tempestivam oportunam, maturam; nam tempore
inoportuno hae caesae arbores cito termites faciunt: ita enim ligni vermes vocantur.
(Tempestivam) nam hac re etiam rustici lunae cursum observare dicuntur; melius enim
ar bores luna decrescente caeduntur. '"Tempestivam: opportune, early; for these trees if
felled at an inopportune time soon make ‘termites’, as vvoodwoims are called.
{Tempestivam:) for in this matter even country folk are said to observe the moon’s course;
for trees are better felled while the moon is waning.”

^The rules for calculating the date of Easter Sunday were adjusted under the auspices of
Theodosius I c. ad 388 consistently with lunar dates here offered, cf H. Leclerq, Diet.
d'Archeologie-chretienne et de Liturgie XIII.2 (1938) 1553—1554, with refs. Cato’s
original lunar observations for cutting timber recommended the moon’s last quarter for
choice, or failing that, the new moon and first quarter, cf. J. Heurgon, "Octavo lanam
Lunam: traces d’une semaine de sept jours chez Varron et Caton”, REL xxv (1947)
236-249, on Cato 37.3^ = Pliny Nat 16.75.194, interpreted 16.74.188 ff.

V. will have Christianized and distorted the source, which was easily misconstrued owing
to its archaism: die bus VII proximis, quibus luna plena fuerit, optime eximetur, meaning
"after the seven days immediately following full moon, it is best cut”, not "on the seven
days”, etc.

^The superfluous mention of the two months named after Julius Caesar and Augustus
immediately after the Christian religion may possibly follow a pattern in which V.
regularly mentions the Emperor after God; cf. I praef, 11.5, 11.18, 111.5.

* tabulae duplices. It is not clear what exactly these were.

^Cf Meiggs, 349; in an emergency there was no option but to build with green timber.

144

VEGETIUS’ EPITOME OF MILITARY SCIENCE

are reported to have clashed, so that these were of six and even more
ranks.* But to the larger warships are attached scouting skiffs,^ having
about twenty oarsmen on each side; these the Britons call f picati?
They are used on occasion to perform descents'* or to intercept convoys
of enemy shipping or by studious surveillance to detect their approach
or intentions.^ Lest scouting vessels be betrayed by white, the sails and
rigging are dyed Venetian blue, which resembles the ocean waves; the
wax used to pay ships’ sides is also dyed. The sailors and marines put
on Venetian blue uniforms too, so as to lie hidden with greater ease
when scouting by day as by night.^

(The art of navigation, ch. 38-43)

38. The names and number of the winds.^

He who sails with an army in an armed fleet ought first to forecast the
signs of storms. For storms and waves have often done greater damage
to warships than hostile attack. In this connexion all the wisdom of
natural philosophy should be applied, for the nature of the winds and
tempests is deduced by studying the skies. So far as the roughness of the
sea is concerned, as caution protects the provident, so carelessness

'Flor. 2.21.5-6 reports that Octavian’s ships had up to six ranks of oars, Antony’s from
six to nine. Cf IV.33.

^scaphae exploratoriae.

'The word is garbled in the MSS. Since the ships were payed with blue wax (as well as
pitch and resin, cf. IV.44) they were in effect painted blue, so some such word as
pictae —“painted”—may be right. Steweehius conjectured *pmcae as a putative plural
latinized form of the M.Du., MLG, LG and mod. German and English word “pinke,
pink”. On the scapha cf. Redd^, 130, Caes. Bell. Civ. 3.24, Bell. Gall. 4.26.4 speculatoria
navigtay Livy 36.42.8.

^superventus.

^The counterpart of the use of scouts on land, cf 111.6. Tactical intelligence is meant, not
strategic, of which V. is largely innocent.

^’The blue disguise is uniquely attested here. Britain was still notoriously the country of
blue woad, cf Claud, de hello Get. 417-418, pan. de Hon. cons. Ill 54, de cons. Stil.
2.248. But the province was also very exposed to sea-borne barbarian raiders and will
have needed special naval surveillance vessels such as these. However, they were not
unique to British waters.

^IV.38-42 derive from Varro’s “//6n navales" (cited at IV.41), whether de Ora Maritima
(lost), so R. Reitzenstein, “Die geographischen BUcher Varros”, Hermes xx (1885) 514 ff.,
and Nielsen (see below), or Ephemeris Navalis (lost), so RE Suppl. VI (1935) col.
1252-1253 (H. Dahlmann), and E. Courtney (ed.). The Fragmentary Latin Poets (Oxford
1993) 246, or de Aestuariis (lost), mentioned in de ling. lat. 9.26.

BOOK IV: SIEGE AND NAVAL WARFARE

145

drowns the negligent. Thus the art of navigation ought first to look into
the number and names of the winds.

The ancients thought that following the position of the cardinal points
only four principal winds blew from each quarter of the sky, but the
experience of later times recognizes twelve. Their names we have set
out in Latin as well as Greek for the avoidance of doubt, so that having
stated the principal winds we shall indicate those adjacent to them on
the left and right.' So let us take our beginning from the vernal
equinox,^ that is, from the eastern cardinal point, whence arises a wind
called Apheliotes, that is, Subsolanus (E. wind); adjacent to him on the
right is Caecias or Euroborus (ENE wind), and on the left Eurus or
Vulturnus (ESE wind). Notus, that is, Auster (S. wind), occupies the
southern cardinal point; adjacent to him on the right is Leuconotus, that
is, Aibus Notus (SSE wind), and on the left Libonotus, that is, Corus
(SSW wind). Zephyrus, that is, Subvespertinus (W. wind), holds the
western cardinal point; adjacent to him on the right is Lips or Africus
(WSW wind), and on the left lapyx or Favonius (WNW wind). The
northern cardinal point is allotted to Aparctias or Septentrio (N. wind);
close to him on the right is Thrascias or Circius (NNW wind), and on
the left Boreas, that is, Aquilo (NNE wind).

These winds are accustomed to blow separately, sometimes in pairs,
but in great storms even three at once. Under their attack the seas,
which are naturally tranquil and quiet, rage with boiling waves. By their
breath according to the nature of the season and location fair weather is
restored after storms and fine conditions are turned to stormy again.
When there is a following wind the fleet reaches its desired port, but an
adverse wind compels it to stand at anchor or go back or risk danger.
And so he is rarely shipwrecked who makes a thorough study of the
science of winds.

‘The original source of this 12-point wind-scale was Timosthenes; Poseidonios took it
over and Varro incorporated this version in his work, from whence V. appears to have
derived it. Cf K.. Nielsen, “Remarques sur les noms grecs et latins des vents et des
regions du ciel”, C&M vii (1945) 105—108, L. Edelstein, I.G. Kidd, Posidonius, fragments
(Cambridge 1972) fr. 137a, theory of winds (Strabo 1.2.21), comm. II p.519 (P. has in
mind the 12-point compass-card of Timosthenes), fr. 138, theory of tides (cf IV.42), frs.
214-219, tidal and coastal waters (cf IV.43).

^v.l. The “vulgate” reads solstitio, “solstice”, which if genuine is V.’s mistake.

146

VEGETIUS’ EPITOME OF MILITARY SCIENCE

39. In which months it is safe to sail.

The next question is to consider months and dates. For the violence and
roughness of the sea do not permit navigation all the year round, but
some months are very suitable, some are doubtful, and the rest are
impossible for fleets by a law of nature. When Pachon' has run its
course, that is, after the rising of the Pleiades, from six days before the
Kalends of June (i.e. 27th May) until the rising of Arcturus, that is,
eighteen days before the Kalends of October (i.e. 14th September),
navigation is deemed safe, because thanks to the summer the roughness
of the sea is lessened. After this date until three days before the Ides of
November (i.e. 11th November) navigation is doubtful and more
exposed to danger, as after the Ides of September (i.e. 13th September)
rises Arcturus, a most violent star, and eight days before the Kalends of
October (i.e. 24th September) occur fierce equinoctial storms, and
around the Nones of October (i.e. 7th October) the rainy Haedi, and five
days before the Ides of the same (i.e, 11th October) Taurus. But from
the month of November the winter setting of the Vergiliae (Pleiades)
interrupts shipping with frequent storms. So from three days before the
Ides of November (i.e. 11th November) until six days before the Ides
of March (i.e. 10th March) the seas are closed. The minimal daylight
and long nights, dense cloud-cover, foggy air, and violence of winds
doubled by rain and snow not only keep fleets from the sea but also
traffic from making journeys by land. But after the birthday, so to
speak, of navigation^ which is celebrated with annual games and public
spectacles in many cities, it is still perilous to venture upon the sea right
up to the Ides of May (i.e. 15th May) by reason of very many stars and
the season of the year itself—not that the activities of merchants cease,
but greater caution should be shown when an army takes to the sea in

'The word is garbled in the MSS, pachnitae, phagnitae, phaeniiae; T. Mommsen, “Zu
Vegetius”, Hermes \ (1866) 131, restored Pachone, the Egyptian month corresponding to
26th April-25th May. Why V. should cite an Egyptian month is unclear, but it may have
been linked in his source to Isis Pelagia, for example (see below). For Pachon, cf. A.E.
Samuel, Greek and Roman Chronology^ Handbuch der Altertumswissenschaftl.7 (Munich
1972) 177. The true resolution of the corruption may be Pachone itaque\ cf IV.38 a
verno itaque solstitio, id est... with Pachone itaque decurso, id est...

^Navigium Isidis or ploiaphesia, a nautical festival held on StlTMarch in honour of Isis
Pelagia, cf J. Roug^, Recherches sur Vorganisation du commerce maritime en
Mediterranee (Paris 1966) 33, Apul. Met. 11.8-17, RE IX (1916) 2084-2132 (Roeder),
M.R. Salzman, On Roman Time: the codex-calendar of 354 and the rhythms of urban life
in late antiquity (Univ. of California 1990) 173. V. is coy about naming the pagan
goddess, whose festival continued to be celebrated even in the sixth century AD, cf. loh.
Lydus de Mens. 4.45.

BOOK IV: SIEGE AND NAVAL WARFARE

147

warships than when the enterprising are in a hurry for their private
profits.*

40. How the signs of storms should be astronomically observed.

Also the rising and setting of other stars provoke very violent storms.
Although fixed dates are appointed for them by the testimony of writers,
nevertheless, since some details change on various occasions and, as
must be admitted, the human condition prevents full knowledge of
heavenly causes, they divide the results of nautical astronomical
observation in three ways. It has been established that storms occur
either about the appointed day, or before or after it. Hence they say in
Greek terms that those occurring beforehand “storm before”,^ those
arising on the regular date “storm’V and those following after “storm
after”But to list them all by name would seem either unnecessary or
tedious, since very many authors have expressed in full the catalogue of
both months and days. The transitions, also, of the stars they call
planets, when they enter or leave signs of the zodiac by the course
prescribed by the will of God the Creator, often tend to disturb fine
weather.^ The days of new moons too are filled with storms and are
very much to be feared by navigators, as is understood not only by
scientific study but the experience of the common people.

41. On signs of the weather.^

There are many signs, too, whereby storms are indicated after fair
weather and fair weather after storms, which the moon’s orb shows as
though in a mirror. A red hue announces winds, blue, rain, a mixture of
both rain-storms and furious squalls. A smiling and bright orb gives

'The greed of merchants which makes them brave the high seas was a commonplace; cf.
Juv. 14.288-302, Ambr. de Off. 1.243, de Elia 70-71. According to CTh 13.9.3 (AD 380)
the seas were open to merchant shipping from the Kalends of April (i e. 1st. April) until
the Ides of October (i.e. 15th. October).

^7cpoxei|iot ^Eiv.

^Xeipd^Eiv.

'’peTaxEipdi^eiv. These terms will have been transmitted by Varro. Cf Pliny
Hist. Nat. 18.57.207, probably from Varro; Paul. Fest. 249.22 L. has the translation
praesiderare for TipoxEipcti^eiv: Praesiderare dicitur cum maturius hiberna tempestas
movetur, quasi ante sideris tempus. Praesiderare is said when a winter storm arises early,
as if before the time of the star.”

^V. alludes to the pagan belief that the planets were gods, showing his Christian
allegiance again.

^'prognostica. Based largely on Verg. Georg. 1.393-463.

148

VEGETIUS’ EPITOME OF MILITARY SCIENCE

promise of fair weather for shipping which it wears on its face,
particularly if at the fourth rising it is not red with the horns blunted or
darkened by a covering of vapour. The sun is also significant when it
rises or closes the day—whether it rejoices in evenly distributed light or
is variegated by cloud in the way, whether it is brilliant with its wonted
brightness or fiery from being beset by winds, or whether it is not pale
or blotched by impending rain. The air, too, the sea itself, and the size
and shape of clouds instruct attentive sailors. Some information is
indicated by birds, and some by fishes, which Vergil included in his
Georgies with almost divine skill,' and Varro studiously developed in
his naval books.^ If pilots profess to know this they do so only insofar
as they have learned it by trial and error; no deeper knowledge has
informed them.

42. On tidal waters, that is, ebb and flow.

The element of the sea is the third part of the world, and in addition to
the blast of the winds it is quickened by its own breath and motion. For
at appointed times, both day and night, there is a surge, which they call
a “flow”,^ which runs forward and back and like rivers in flood now
overflows onto land, now flows back into its own depths. This two-way
reciprocal motion helps ships on their course when it is behind them,
and retards them when in front. For one intending to do battle it is a
thing greatly to be avoided. The force of a tide is not overcome by the
help of oars, for even the wind yields to it on occasion. And since in
different regions it varies at appointed times according to a different
state of the moon’s waxing and waning, he who is going to fight a
naval battle ought to find out the characteristics of the sea and locality
before any encounter.

'Verg. Georg. 1.351 ff.

^On Varro’s '‘libri navales'\ cf. IV.38 note. The information indicated by fishes is not
in Vergil, nor in Aratus’ ''Diosemeid' (the last 400 lines of his Phaenomena), on which
Vergil drew, nor in surviving parts of Latin poetic adaptations of the same by Cicero,
Varro Atacinus, Germanicus and Avien(i)us. But it is in Pliny Hist. Nat. 18.361, from
Varro, probably; see ibid. 18.348-350 for a quotation from Varro on weather-signs from
the moon and Is id. de Nat. 38 (p.303 Fontaine) for similar from the sun.

^rheuma, Greek for “tide”.

BOOK IV: SIEGE AND NAVAL WARFARE

149

43. On navigational knowledge and oarsmen.’

It is the responsibility of sailors and pilots to acquaint themselves with
the places in which they are going to sail and the harbours, so as to
avoid dangerous waters with projecting or hidden rocks, shallows and
sandbanks. Safety is greater, the deeper the sea is. In navarchs^ close
attention is required, in pilots^ skill, and in oarsmen'’ strength, because
naval battles are staged in a calm sea, and the massive warships strike
through the enemy with their “beaks”^ and avoid their attack in turn not
by means of the breath of winds but by the beat of their oars. In this
operation the muscle of the oarsmen and the skill of the officer who
guides the rudder win a victory.

(Naval warfare, ch. 44-46)

44. On naval weapons and torsion-engines.

Land warfare requires many types of arms; but naval warfare demands
more kinds of arms, including machines and torsion-engines as if the
fighting were on walls and towers. What could be crueller than a naval
battle, where men perish by water and by fire? Therefore protective
armour should be a particular concern, so that soldiers may be protected
with cataphracts, cuirasses, helmets and also greaves. No one can
complain about the weight of armour, who fights standing on board
ships.^ Stronger and larger shields are also taken up against the impact
of stones. Besides drags’ and grapnels* and other naval kinds of
weapons, there are arrows, javelins, slings, “sling-staves”, lead-weighted

'The subject makes it likely that this chapter is derived from Varro de Ora Maritima also,
or at least the first part is.

^navarchi, i.e. captains, cf. IV.32.

^gubernatores.

^remiges.

^rostra, i.e. rams.

^*Cf. 1.20. V. is still anxious about the alleged unpopularity of heavy armour; quite likely
he has simply transposed the situation on land to that at sea. But the point is well-taken
that there is no role for light-armed troops on board ship, because of the static conditions
of close-quarters combat at sea. Cf Leo Tact. 19.13 for the continuing influence of this
chapter on ninth-century Byzantine tactici.

^falces.

^harpagones. Cf Front. Strat. 2.3.24, Polyb. 1.22-23 (KopaKcg = com, ‘"crows”), Flor.
1.18.8-9, Zonar. 8.11 for the introduction of the harpago or manus ferrea, “iron hand”,
under C. Duellius in 260 BC during the First Punic War, and the stopping of fluid naval
battle-tactics to facilitate stationary boarding operations.

150

VEGETIIJS EPITOME OF MILITARY SCIENCE

darts, mangonels, catapults, and hand-catapults,' shooting darts and
stones at each side. More dangerously still, those confident of their
courage move up their warships alongside, throw out bridges and cross
over to the enemies’ ships to fight it out there with swords hand-to-
hand, or comminus as it is called. On larger warships they even erect
fortifications and towers, so that they may more easily wound or kill
their enemies from higher decks as if from a wall. Arrows wrapped in
“burning-oil”, tow, sulphur and bitumen are planted blazing by
catapults^ in the bellies of hostile ships, and soon set light to the
planking payed with wax, pitch and resin—so much kindling for fires.^
Some are slain by steel and stones, others are forced to burn to death in
the water. Among so many forms of death the bitterest fate is that the
bodies are left unburied to be eaten by fish.''

45. How ambushes are set in naval warfare.

Just as in land warfare, descents are made upon sailors who are
unsuspecting, and ambushes are laid about suitable narrows between
islands. This is done with the idea of destroying them more easily, being
unprepared. If enemy sailors are weary from lengthy row ing, if pressed
by head-winds, if the tide is flowing against the ships’ “beaks”, if the
enemy are asleep suspecting nothing, if the anchorage that they hold has
no (other) exit, if a desired opportunity for battle occurs, one should
take Fortune’s favour in one’s hands and give battle at the opportunity.^
But if thanks to his caution the enemy avoids ambush and gives battle
in a general engagement, the lines of the warships should be drawn up
not straight as in a land battle, but curved like the moon, so that the
wings are brought forward and the middle of the line is bowed. Then,
if the adversaries try to break through, they may be surrounded by the
formation itself and sunk. The chief strength of both warships and
marines is placed on the wings.^

^scorpiones, i.e. manuballistae, cf IV.22.

^Cf, IV. 18 for fire-darts.

^Cf. IV.37 for paying ships’ sides.

'Cf Claud, paw. de Hon. cons. /K 628-629, Ovid. Tr 1.2.53-56.
^V. transfers to a naval context 111.19 init.

"So Philo Byz. Polior 104.

BOOK IV: SIEGE AND NAVAL WARFARE

151

46. What to do when naval battle is joined in an open engagement.

Moreover it is advantageous for your fleet always to use the deep, open
water, while the enemy’s is pushed inshore, because those who are
thrust towards land lose the speed to attack.^ In this type of encounter
it has been established that three kinds of weapon contribute most
towards victory, beams, drags and battle-axes. The '‘beam”^ is the name
for a thin, long shaft like a yard-arm, which hangs from the mast and
has an iron head at both ends. When ships attach themselves to the
adversaries’ starboard or port sides, they violently drive this instead of
a battering-ram. It is a sure method of cutting down and killing enemy
marines and sailors, and often pierces the ship itself^ The “drag”"* is the
term for a very sharp iron blade curved like a sickle and mounted on
long poles. It quickly cuts the rigging^ from which the yard-arm is
suspended, and the sails collapse rendering the warship slow and useless.
The “battle-axe”^ is an axe which has on both sides a very broad and
very sharp iron blade. By means of these, in the midst of the heat of the
battle very skilled sailors or marines in small skiffs^ secretly cut the
cables binding the adversaries’ steering-gear. Once this is done the ship
is captured immediately, being disarmed and disabled. For what escape
is left for him who has lost his rudder?

On the subject of river patrol-boats,^ which guard outposts on the
Danube with daily watches,^ 1 feel I should keep quiet, because their

'Cf. Polyb. 1.51.4 IT.; F. Lammert, KUo xxxiii (1940) 278.

^asser

W. alone describes the asser as a naval aries. Otherwise it is the pole on which a
harpago or grapnel is mounted, cf. Livy 30.10.16, Appian Bell. Civ. 5.118 (491)-119
(496), Q. Curt. 4.3.24-25, Caes. Bell. Gall. 3.14.5-7.

*falx, literally “sickle”.

^chalatorii. Lang rightly brackets sub. furies after Keller as a scribal gloss, subaudi
Junes —“understand ropes”. On V.’s use of the Greek word chalatorius, cf. chalo at IV.23.

^'bipennis, battle-axe, double axe.

^scaphulae. Presumably the same as at IV.37; so Redde, 130.

^lusoriae.

"Cf. CTTi 7.17 (AD 412).

152

VEGETIUS’ EPITOME OF MILITARY SCIENCE

increased use has discovered a more advanced science for them than
ancient theory had to show.'

FINIS.

‘So too at III.26 V. refused to discuss cavalry, modem advances having made the book*
doctrines obsolete. The Epit. was not intended to be a complete Art of War, but a critique
of current failings and strategies. In view of V.’s fondness for prologues and epilogues,
book IV ends abruptly; comparison with 111.26 suggests the possibility that a final
epilogue may have fallen out of the tradition here. However, as the naval section is
presented in the ancient Synopsis as an appendix, and is given short shrift in the prologue
at IV.31, it may well be that V. did not consider another epilogue needful after IV.30.

Abbreviations and Bibliography

[Standard abbreviations for journals used by L *annee philologique Iviii

(1987) are copied or slightly adapted in the commentary.]

ANRW H. Temporini et al. (edd.), Aufstieg und Niedergang des
romischer Welt (Berlin 1972- ).

BAR British Archaeological Reports (Oxford).

BGU Berliner Griechischer Urkunden I—IV, Aegyptische Urkunde aus
den koeniglichen Museen zu Berlin (Berlin 1895-1912).

CCSL Corpus Christianorum Series Latina (Tumhout, Belgium).

Chron. min. T. Mommsen (ed.), Chronica minora, Monumenta
Germaniae Historiae Auctorum Antiquissimorum l-III (Berlin
1892-1898).

CIL T. Mommsen et al. (edd.). Corpus inscriptionum Latinarum
(Berlin 1863-).

CJ P. Krueger (ed.), Codex Justinianus, (Berlin 1877).

CSEL Corpus scriptorum ecclesiasticorum Latinorum (Vienna).

CTh T. Mommsen (ed.). Codex Theodosianus (Berlin 1905).

Dig. T. Mommsen (ed.), Digesta lustiniani Augusti (Berlin 1868-70).

edn. edition.

Epit. Epitoma Rei Militaris.

FGH F. Jacoby (ed.), Die Fragmente der griechischen Historiker
(Berl in-Leiden 1923-57).

JRMES Journal of Roman Military Equipment Studies

lit. literally.

MAMA W.M. Calder et al. (edd.), Monumenta Asiae Minoris Antiqua
I- (London 1928- ).

Mul. Digesta Artis Mulomedicinae.

NCO Non-commissioned officer.

Nov. Th. II P.M. Meyer (ed.). Leges novellae Theodosii II Augusti
(Berlin 1905).

Nov. Val. Ill P.M. Meyer (ed.), Leges novellae Valentiniani III Augusti
(Berlin 1905).

OCD Oxford Classical Dictionary (1970^).

O. Douch H. Cuvigny, G. Wagner (edd.), Les Ostraca grecs de Douch,
Inst. fran9ais d’arch. orient. XXIV fasc. 1.1-57 (Cairo 1986), fasc.
11.58-183 (Cairo 1988).

OLD Oxford Latin Dictionary (1968-82).

P. Lips. L. Mitteis (ed.), Griechische Urkunden der Papyrussammlung
zu Leipzig (Leipzig 1906).

154

VEGETIUS’ EPITOME OF MILITARY SCIENCE

P. Monac. A. Heisenberg, L. Wenger (edd.), Veroeffentlichungen aus
der Papuryssammlung der K. Hof- und Staatshibliothek zu Munchen:
Byzantinische Papyri (Leipzig-Berlin 1914).

P. Oxy. B.P. Grenfell, A.S. Hunt et al. (edd.). The Oxyrhynchus Papyri
(London 1898- ).

PLRE A.H.M. Jones, J.R. Martindale, J. Morris (edd.), Prosopography
of the Later Roman Empire 1 (Cambridge 1971), J.R. Martindale
(ed.), vol. II (1980).

RE Pauly-Wissowa (edd.), Reai-Encyclopadie der classischen
Altertumswissenschaft (Stuttgart 1894- ).

ref. reference.

S. O. Seeck, Regesten der Kaiser und Pdpste (Stuttgart 1919).

SB F. Preisigke et al. (edd.), Sammeibuch griechischer Urkunden aus
Agypten I-(Strasbourg 1915- ).

SEG J.J.E. Hondius et al. (edd.), Supplementum epigraphicum Graecum
I- (Leiden 1923- ).

SHA Script ores Historiae Augustae.

Suda A. Adler (ed.), Suidae lexicon (Teubner edn., 1928-38).

TLL Thesaurus Linguae Latinae.

V. Vegetius.

v.l. varia lectio.

Select Bibliography

[A full Vegetius bibliography was published by R. Sablayrolles in 1984,
and the general bibliography of the Roman army is truly vast.
Accordingly only a representative selection of the chief works cited in
the introduction and commentary are listed, together with a few other
recent books and articles which seemed relevant.]

Alfbldy, G., Romische Heeresgeschichte: Beitrdge 1962-1985,
Mavors Roman Army Researches III (Amsterdam 1987).

Anderson, A.S., “The imperial army”, in J. Wacher (ed.). The Roman
World I (London-New York 1987) Pt. 3.

Andersson, A., Studia Vegetiana: commentatio academica (Uppsala
1938).

Baatz, D., Bauten und Katapulte des romischen Heeres, Mavors
Roman Army Researches XI (Stuttgart 1994).

Birley, E., The Roman Army, Papers 1929—1986, Mavors Roman

ABBREVIATIONS AND BIBLIOGRAPHY

155

Army Researches IV (Amsterdam 1988).

Bishop, M.C., Coulston, J.C.N., Roman Military Equipment from the
Punic Wars to the Fall of Rome (London 1993).

Blockley, R.C. (ed.), The Fragmentary Classicising Historians of the
Later Roman Empire: Eunapius, Olympiodorus, Priscus and Malchus
I-II, ARCA classical and mediaeval texts, papers and monographs, 6,

10 (Liverpool 1981-83).

Breeze, DJ., Dobson, B., Roman Officers and Frontiers, Mavors
Roman Army Researches X (Stuttgart 1993).

Campbell, B., “Teach Yourself how to be a General”, JRS Ixxvii
(1987)13-29.

Campbell, B., The Roman Army, 31 BC~AD 337, A Sourcebook
(London 1994).

Campbell, J.B., The Emperor and the Roman Army 31 BC-AD 235
(Oxford 1980).

Cichorius—references are to the re-edition of C. Cichorius’ plates by
F. Lepper, S. Frere, Trajan's Column (Gloucester 1988).

Clarke, Lt. J., Military Institutions of Vegetius (London 1767).

Clauss, M., “Ausgewahlte Bibliographie zur lateinischen Epigraphik
derromischen Kaiserzeit (1.-3. Jh.)”, ANRW 11.1 (1974) 840-847 on the
Roman army.

Davies, R.W., Service in the Roman Army (Edinburgh 1989).

Emout, A., Meillet, A., Dictionnaire Etymologique de la langue
latine, augmented edn. by J. Andre (Paris 1979"^).

Feug^re, M., Les armes des romains de la Repuhlique a I’Antiquite
tardive (Paris 1993).

Fink, R.O. Roman Military Records on Papyrus, Philological
monographs of the American Philological Association XXVI (Cleveland,
Ohio 1971).

Fomi, G., Esercito e Marina de Roma Antica, Mavors Roman Army
Researches V (Stuttgart 1992).

Garlan, Y., “Le Livre ‘V’ de la Syntaxe mecanique de Philon de
Byzance, Texte, Traduction et Commentaire”, Recherches de
Poliorcetique grecque, Bibliotheque des Ecoles fran^aises d’Ath^nes et
de Rome CCXXIII (Paris 1974) 279-404.

Gilliam, J.F., Roman Army Papers, Mavors Roman Army Researches

11 (Amsterdam 1986).

Goffart, W., “The Date and Purpose of Vegetius’ De Re MilitarC,
Traditio xxxiii (1977) 65-100 = id., Rome's Fall and After (London
1989) 45-80.

156

VEGETIUS’ EPITOME OF MILITARY SCIENCE

Goodbum, R., Bartholomew, P. (edd.), Aspects of the Notitia
Dignitatum, BAR Supp. Ser. 15 (Oxford 1976).

Grosse, R., Romische Militargeschichte von Gallienus bis zum Beginn
der byzantinischen Themenverfassung (Berlin 1920).

Hassall, M.W.C., Ireland, R.I. (edd.), De Rebus Bellicis: Papers
presented to Prof E.A. Thompsony BAR Int. Ser. 63 (Oxford 1979).

Hoffmann, D., Das spatromische Bewegungsheer und die Notitia
Dignitatuniy Epigraphische Studien, VII. 1-2 (Dusseldorf 1969),

Jones, A.H.M., The Later Roman Empire (Oxford 1964).

Jordan, H., M. Catonis praeter librum de re rustica Quae Extant
(Leipzig 1867).

Kempf, J.G,, “Romanorum sermonis castrensis reliquiae collectae et
illustratae”, Jahrbucher fur klassische Philologie, Suppl. XXVI (1901)
338-400.

Keppie, L.J.F., The Making of the Roman Army: from republic to
empire (London 1984).

Kolias, T.G., Byzantinische Waffeny ein Beitrag zur byzantinischen
Waffenkunde bis zur lateinischen Eroberung, Byzantina Vindobonensia
bd. XVII, Osterreichische Akadamie der Wissenschaften (Vienna 1988).

Lang, C. (ed.), Vegetii Epitoma Rei Militaris (Teubner edn., Leipzig
1885').

Le Bohec, Y., The Imperial Roman Army (Paris 1989, tr. London
1994).

Lenoir, M., Pseudo-Hygin, Des Fortifications du Camp (Bude edn,,
1979).

Lepper, F., Frere, S., Trajan's Column (Gloucester 1988).

Liebeschuetz, J.H.W.G., Barbarians and Bishops: Army, Church, and
State in the Age of Arcadius and Chrysostom (Oxford 1990).

Lommatsch, E. (ed.), Vegetii Digest a Artis Mulomedicinae (Teubner
edn., Leipzig 1903).

Luttwak, E.N., The Grand Strategy of the Roman Empire
(Baltimore-London 1976).

MacMullen, R., “How big was the Roman Imperial Army?”, Klio Ixii
(1980) 451-460.

Marquardt, J., Romische Staatsverwaltung II (Leipzig 1884').

Marsden, E.W., Greek and Roman Artillery: Historical Development
(Oxford 1969).

Marsden, E.W., Greek and Roman Artillery: Technical Treatises
(Oxford 1971).

Milner, N.P., Vegetius and the Anonymus De Rebus Bellicis, D.Phil.

ABBREVIATIONS AND BIBLIOGRAPHY

157

thesis, Oxford 1991.

Neumann, A.R., RE Suppl. X (1965), s.v. “Vegetius”, col. 992-1020.

Onnerfors, A. (ed.), P, Flavii Vegeti Renati Epitoma rei militaris
(Teubner edn., Stuttgart 1995).

Phillips, T.R. (ed.), Flavius Vegetius Renatus, the Military Institutions
of the Romans (Harrisburg, Pa. 1944), repr. Westport 1985.

Redd^, M., Mare Nostrum: les infrastructures, le dispositif et
rhistoire de la marine militaire sous Fempire Romain (Rome 1986).

Ruggiero, E. de (ed.), Dizionario Epigrafico de Antichita romane I-
(Rome 1886-).

Sabbah, G., “Pour la datation thdodosienne du De Re Militari de
V^gdce”, Centre Jean Paleme, Memoires II (Univ. de Saint-Etienne
1980), 131-155.

Sablayrolles, R., “Bibliographic sur Vepitoma rei militaris de
V^g^ce”, CGRAR III (1984) 139-146.

Schenk, D., Flavius Vegetius Renatus: Die Quellen der Epitoma rei
militaris, Klio, Beitrage zur alten Geschichte XXII (N.F. IX) (Leipzig
1930), repr. Nuremberg 1963.

Speidel, M.P., Roman Army Studies I (Amsterdam 1984), II (Stuttgart
1992), Mayors Roman Army Researches I and VIII.

Speidel, M.P., Riding for Caesar, The Roman Emperors* Horse
Guards (London 1994).

Stelten, L.F., Flavius Vegetius Renatus: Epitoma Rei Militaris, edited
with an English translation (New York, Bern, Frankfurt, Paris 1990).

Tomlin, R.S.O., “The army of the late Empire”, in J. Wacher (ed.).
The Roman World I (London-New York 1987) Pt. 3.

Walde, A., Hofmann, J.B., Lateinisches Etymologisches Worterbuch
(Heidelberg 1938-54^).

Watson, G.R., “The army of the Republic”, in J. Wacher (ed.). The
Roman World \ (London-New York 1987) Pt. 3.

Watson, G.R., The Roman Soldier (London 1969).

Webster, G., The Roman Imperial Army of the first and second
centuries AD (London 1985^).

Wheeler, E.L., Stratagem and the vocabulary of military trickery,
Mnemosyne suppl. CVIII (Leiden 1988).

Whitehead, D., Aineias the Tactician (Oxford 1990).

Index of Gods^ People and Places

Achaia 140

Actium 141, 143

Africa 90, 110, 112, 140

African 2

Alan 19, 119

Albinus 90

Allied 65

Allies 30, 34

Ambrones 90

Antiochus 112

Antony 141

Athenian 9, 27, 62

Augustus 1, 10, 26, 37, 141

Balearic Isles 16

barbarian 17, 22, 23, 29, 31, 51, 72, 86, 89, 128, 140
Bessi 43, 135
Britons 144

Caepio 90
Campania 141
Campus Martius 12
Capaneus 133
Capitol 120, 125, 137
Carthaginian 62, 99
Cato 9, 14, 15, 33
Celsus 9
Celtiberian 31
Christ 35
Cimbri 90
Cincinnatus 4
Circus 57
Claudius 15
Crete 140
Cyprus 140

Dacian 27
Dalmatia 141
Danube 151

INDEX OF GODS, PEOPLE AND PLACES

159

Dardanian 31
Darius 64
Diocletian 17

Egypt 140

Emperor 1, 2, 19, 27, 29, 34, 35, 37, 41, 56, 52, 59, 62, 71, 119,
120, 140, 141

Empire 2, 8, 9, 27, 86, 120, 140
Epiros 140
Epirote 27

Fortune 150
Frontinus 9, 33

Gaul 2, 31, 90, 137, 140
German 2

God I, 35, 51, 71, 147
Goth 19
Gratian 19

Greek 3, 9, 31, 145, 147

Hadrian 10, 26
Hannibal 28, 63, 112
Herculiani 17
Holy Spirit 35
Homer 6
Hun 19, 119

ladera 141
Illyricum 17
India 27
Indian 113, 119
Ionian Sea 141
Italy 63

Joviani 17
Jugurtha 113

Libumi 141
Libumia 141

160

VEGETIUS’ EPITOME OF MILITARY SCIENCE

Lucania 112, 113

Macedonia 140
Macedonian 27, 31, 62
Mallius 90, 137
Marius 90
Mars 27
Marsian 27
Mattiobarbuli 17
Mauretania 140
Maximian 17
Mazices 110
Metellus 90
Minotaur 74
Misenum 140, 141
Mithridates 64, 112
Moesian 27

Numantia 89
Numantines 15
Numidia 113

Octavian 1
Oriens 140
Orient 113

Paeligni 27
Patemus 9
Persian 27, 89, 119
Pompeius II
Pontus 140
Propontis 140
Punic war 28
Pyrrhus 112

Ravenna 140, 141
Regulus 62
Rhodes 132

Roman 2, 8, 19, 27, 30, 32, 35, 62, 65, 91, 112
Roman People 2, 9, 15-16, 65, 140

INDEX OF GODS, PEOPLE AND PLACES

Romans 2, 4, 11, 13, 27, 28, 31, 55, 62, 86, 89, 99, 112
Rome 1, 120, 125, 137
Romulus 37

Sallust 5, 11
Samnite 27
Saracen 119
Sardinia 140
Scipio 107

Scipio Africanus 15, 89
Sertorius 8, 11
Sicily 140
Spain 89, 140
Spaniard 2

Spartan 9, 27, 62, 63, 86, 99

Teutones 90
Thebans 133
Thessalian 27
Thracian 27
Tiber 4, 12
Trajan 10, 33
Tydeus 6

Urcilliani 110

Varro 148

Vergil 18, 148, “Mantuan” 7
Vespasian 37

Xanthippus 62
Xerxes 64

THE ONLY LATIN ART OF WAR TO SURVIVE, Vegetius’
Epitome was for long an essential part of the medieval prince’s
military education. The core of his proposals, the maintenance of
a highly-trained professional standing army and navy, was
revolutionaiy for medieval Europe, while his theory of
deterrence through strength remains the foundation of modem
Western defence policy. This annotated translation highlights the
significance for his own age of Vegetius’ advice, written just
before the fall of the Roman Empire in the West, at a time when
economic weakness and political disintegration threatened to
undermine the strategic defensive structure that had underpiimed
the Roman State for so long.

The main thrust of his reforms is to confront the problems of the
fragmentation of the army, the barbarization of its personnel, the
loss of professional skills, and the substitution of mercenaries for
standing forces. The accent of the work is on the practicalities of
recruiting and training new model armies (and navies) starting
from scratch, and on the strategies appropriate to their use
against the barbarian invaders of the period.

N. P. MILNER took First Class Honours in Greats at Christ Church,
Oxford, in 1982. After working for a time in the City and as a
schoolmaster, he returned to Oxford to research^egetius for a DPhil
thesis. He has also been active in archaeological and epigraphical
suivey work in S.W. Turkey undertaken at ancient sites by the British
Institute of Archaeology at Ankara, and has worked on the epigraphical
Nachlass of the late Alan Hall.

Liverpool
University
tess

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Translated Texts for Historians"

See other formats